It's pitiful that any company would do such a thing. I've got a Cosori air fryer which I'd love to fix but the markings on the chip that has failed have also been lasered away!
I have the same yellow component tester since 2017. I wanted to have as you said, something to identify a component as good/bad and if good, how good it is (ESR, hfe, etc). After using it for many years, I get to trust it . Last year, for my birthday, I got another component tester, TC1. The main differences are the color display, the plastic case and the rechargeable battery. As functionality, the TC1 seems to give me nore details about transistors, but its measurements are in par with the old one, maybe some 5-10% difference. However, now I can not swear that a resistor is 220 or 226 Ohm. What I say is that you should trust the one that give you the most consistent values when test a component; I see that your DSO-TC3 gives different values at each testing, that hints to a error in firmware or in the schematic or the comonents they used.
You mean the DSO-TC3 in the video? I've been using the yellow one for quite some time and as you say, I learnt to trust it, I never felt it was giving me inaccurate results. Of course, as I said, I've always been looking for "good" "bad" outcome, nothing more.
I had bought this meter from an estate sale for pennies and was noticing that it was giving off readings as well. I'm glad these two videos came up on your channels Tony.
Thanks Tony, I use the 100Hz frequency as I find it results in a clearer result between good and bad caps, higher frequencies result in a narrower spread in ESR between good and bad caps, making it less obvious what their condition is like.
Would have been good to do extra tests at 1khz/10khz to see how much the ESR changed. The TransistorTesters original firmware claims to produce a value more accurate to a 10khz test
So, I'm not surprised that voltage measurement would not see the frequency, as you are measuring the capacitor voltage, and smoothing the voltage is the role of the capacitor. What I'm thinking is measure the current going to the capacitor on the oscilloscope (like voltage drop of a 1ohm resistor in series). I assume you could see current going 2 ways with the measurement frequency.
Those ATMega tester don't work at high frequencies, it's all on the document linked in the description. Anyways, I doubt an electrolytic cap can do much at 100KHz.
The FNIRSI TC3 is a multi function device being primarily an oscilloscope. The component test function is secondary. I use the component test as a go/no go indication and to find out the value of random capacitors. For that it works very well. If I were interested in ESR and other measurements, I would buy a more expensive device with switchable frequencies. You need a high frequency for low pfs and a low frequency for electrolytics. I think maybe you expect too much from the TC3 which, for the price, and it isn't a dedicated component tester, actually works very well. Not many of us have three different devices to compare test results!
I think it's fair to expect when they present a device as "that cheap LCR of the kind you used before, but now with a scope and extra features and much more expensive", that the scope and all those other features might be bad, but at least the LCR functionality doesn't work worse than you're used to! It would have been my assumption for sure.
There is no primary/secondary, there is just an "overall bad". This device aims to be a "I can do everything" device. But as usual: Something able to do everything, does everything more or less bad. I wouldn't recommend either, not the scope function nor the component tester. And in the end, this device got compared to a dirt cheap component tester, but failed horrible. Yeah, meh... I think they tried to port the old AVR code to some kind of ARM device, with completely different I/O port characteristics, but failed to modify the code in a proper way. The original includes a lot of constants and mathematical tricks to get to the point. I assume they tried something, they don't even understand on a basic level. So, it might be fixable with firmware updates, but... as Chinese manufacturer usually handle this, I won't expect any improvement here.
I have 2 of those kind of component testers. They work ok for semiconductors, but for caps, resistors and inductors, I trust my dedicated LCR meters way more. If I want more detailed swept frequency measurements of LCR’s, I use my Analog Discovery with its impedance-measurement attachment. When I worked in a development lab, there I ahd access to very high-ended LCR-bridges, that are unaffordable to a normal mortal on retirement, but the Analog discovery comes close, except if you want a DC bias. Those cheap component testers are great for simple semiconductor tests, but for more detailed tests, curve tracers are the way to go. With all that being said, wish those component testers were available when I started in electronics in the 1970’s.
You are absolutely right, those testers are great for what they do but because they say "ESR" on it it won't mean it's going to be accurate. It's good from a hobbyist perspective as you say as long as we are all aware of the limitations!
@@Tony359_2 It is the sum of the reactance of the capacitance plus the Equivalent Series Resistance plus Equivalent Series Inductive Reactance.Xc + Resr + Xl.If the total impedance at is for example 15 mOhm , you should subtract Xc and Xl to get Resr.The only correct method is a LCR bridge meter who can extract ESR and capacitance correctly at a certain frequency.DER-EE DE-5000 LCR meter is a good precise meter at a reasonable price.
Thank you much for the follow up. ❤ The new meter seems to be a combination of the DSO-138 "Oszilloskop" and the old component tester. But as chinese standard: a bad copy without proper (german) engineering. I have the old tester and now i know to keep it, because the new thing isnt worth any pennys.
@@Tony359_2 For sure! A bit more investigation how and why the germany guy has made his software as it is would come up in a very nice "multitool". But this thing is useless.
It is possible to put "original" firmware into the old-type tester. Don't know for M-Tester, but for original firmware, it is recommended to measure a capacitor across terminals 1 and 3.
There's no hard reason for that i don't think. Other than layout symmetry reasons and tiny little capacitive differences. On the schematic and microcontroller all 3 pins are identical.
And markings have been removed. I bet they use at least some Open Source components for their firmware (if not most of it) and then they have the audacity to do that. Not that it is very difficult to find out what the chip is anyway. BTW. Most of Open Source licenses require to publish the source code (not the binaries) together with any modifications. I wonder if they do that. And I think the original component tester of this kind was entirely Open.
For entering selftesting/calibrating mode on the classic tester you should (according to the original documentation) short pins 1, 2,3, then press the button and press it again within 2 seconds. If this doesn’t work, the function was disabled when compiling the firmware.
Great video Tony, Thank You. I've been wondering about the FNIRSI Products, as I'm re-populateing my workspace after being burgled recently. So, Hyper Tinkering.! Sounds great 😎😆. I don't know why, but today is the first time yt has let come to this channel. There is No Subscribe button, and No Notification button.??
Oh no, I am so sorry to hear that. I hope you get back up and running very soon. I don't know, you should see a subscribe button somewhere. There shouldn't be anything special about this channel, it's just an independent one, I just own both.
@Tony359_2 I'm not sure what changed, but I am Subscribed now 👍😎😆✅ I prefer to build my own gear and equipment, drag all those years of math and theory into the real world. (lol) Maybe this is a hint to update / upgrade. I count myself lucky, that I didn't have any client jobs at the time.
I have an separate esr meter and ive watched it with osc. Can confirm it can test at 10khz and even 100khz. Other bonus is that it has leads and in most cases desoldering isn't needed for caps, it doesn't trigger any semiconductors
If you don't want to test other components and just want a somewhat affordable and perhaps slightly more reliable LCR meter, maybe check out the JYE Tech M162. It's a DIY kit for around 50 Euro, and specializes in just LCR (no semiconductors). But you can manually select what you want to test (no auto-id that goes awry in edge cases). And it'll not only tell you capacitance, but also ESR, impedance, Q and D value, and theta. Testing frequency is selectable between 100 Hz and 1 KHz, so it does use actual frequency testing. JFTR: It gave plausible results during my experiments, but I have not done any extensive, proper testing with expensive, calibrated equipment. So basically: caveat emptor.
thank for that, it looks like a good option. I'm expecting an LCR meter from a small manufacturer - it's apparently a decent one. I could not source the DE-5000, it's impossible to find it here in the UK and the manufacturer could not help. Thanks again!
Basically, the impedance is NOT the same as ESR. Also, the math to calculate the values from assumed equivalent circuit to another - assumed series connection on the ESR or assumed parallel connection of the leakage (!) resistor - involves quite challenging list of formulas. General Radio used to print the collection in their LCR meter manuals. I think I still have one of those manuals, maybe something like GR 1607 or 1608 somewhere. You might find them through Google search, I have not tried. Anyway, I suspect the ATMega or other chip can do conversions (calculations), running with 8 MHz or other high speed clock, while it is probably not trying to make the actual primary measurements at any of the customary frequencies.
Thanks and sorry for assuming Impedance = ESR :) The ATmega is not using any high frequency at all, it's all described in the manual linked in the description but there is no HF involved, it's a very clever approach :)
I'd buy two meters more to divide my certainty by two. Or a cheap oscilloscope with an square wave generator (rectified sine or pure sine would also do) to be certain of how exact the values are. With tons of ´patience, a bridge. Go for it Tony. :)
Excellent testing,, Tony, especially when you go back to the original design to understand how it is measuring capacitance. Maybe a firmware update for the DSO-TC3 will help, but most people will be interested in this unit as a component tester, not a signal generator or oscilloscope. It really needs to be as good if not much better than then the original component tester it is clearly designed to replace.
Indeed, I've had my tester for many years, I thought a more modern replacement could at least match it but no. Someone has mentioned the TC3 is primarily an oscilloscope... I don't understand! :D
It seems that FNIRSI is well-aware of the flaws of this product. Evidently, they've released the new LCR-P1 component tester in the past couple of days to fill this role suitably. There is a thread on EEVBlog forums, but it's recent enough that nobody has it delivered yet. Also, judging from the fact that the listing cleverly refuses to use the term LCR anywhere except the SKU identifier, I'd still be doubtful of them redeeming themselves even on second try.
Not only do you need the test frequency, but you need to be able to change the frequency. Different caps under test require different freqs. You wouldn't test a 3300uF cap at 100KHz and you wouldn't test a 33pF cap at 120Hz. I got a DE-5000 and haven't looked back. I only use these testers for IDing mystery transistors/diodes.
For sure. Though after many years I never felt that old tester ever fooled me. Not accurate by any mean - the original designer confirms that - but it gives you a good idea. I'm trying to get a DE-5000, not easy in the UK apparently thanks to some arguable political decisions with our borders...
@@Tony359_2 Hello, Tony! Great video as always! Now, I've never had LCR meter, so I decided to splurge some money on my New Year gift and got Peak Atlas ESR70 Gold. It's made in UK and I got it because many repair shops recommend it to me and online reviews were also pretty favorable. De ree's was out of option because of the price. Maybe you could check Peak's one out?
Apart from my 1st comment on the possible reasons why the Fnirsi lacks in comparison to the original, and now trying to answer you on the question of the "how" and "what frequency" a capacitor tester operates... Well, capacitor testers usually are designed around two different approaches. One is the "DC" approach or let's call it, the "Charge - Discharge" approach. The other is the "AC approach" or "reactance" approach. Proper LCR meters use the second method and actually apply an frequency to the capacitor (or inductor) and measure its impedance (the complex sum of the resistive and reactive reactances of the component). The mini-testers, however, use the DC approach so... No, they're not using any frequency so to speak. They short the capacitor (to establish a known zero charge starting point) and then they apply a known voltage with a known resistance in series, and use the capacitor charging formula to estimate the capacitor value based on the time it takes for it to charge from zero up to a certain intermediate voltage. This is usually what multimeters with capacitance measurement do... The AVR based tester, being micro-controlled, can do a few more "clever" things, like measuring the voltage at different points in time and, by comparing the actual charge discharge curve to the theoretical exponential curve, it can derive some of the capacitor imperfections. For the ESR, for instance, it first makes an estimation of the capacitance value based on a first "time-to-charge" method... And after it has a baseline, it can then measure it again using, for instance, a different charging resistor, expecting to see a calculated change in the time-to-charge. But when it actually charges in a different time than the calculated, it can infer the ESR and the loss. The method has limited capacities because the small device don't have that many different resistors and charging speeds to obtain an ideal measurement of a large range of capacitors so, it's probably better at measuring those things for a certain range of capacitor values, while for other values, it probably looses resolution amd precison but still can calculate a rounded off value. It's quite a clever method.
Hi Tony, datasheets usually list the worst-case conditions. About the measurement of 5 ohms with the resistor in series, in my opinion, maybe because that's its limit. Maybe by design and because it would be a damaged component, it doesn't measure more than 5 ohms. Anyway, it's useful.
Despite the appearance, the old and rather amateur looking one is actually an exact implementation of the original design by the german guy who had the original idea. Even though it had its limitations, the designer did quite an extensive job in perfecting it and correcting for its shortcomings. Cut many years past, to the FNIRSI product, the concept (of the original design) was quite known already and replicated and modified in subsequent implementations to a point that many people and many companies felt "like" having "mastered" the original concept but... As with anything involving low level programming, it goes a great distance between one "thinking" he understood what a program does and one actually understanding it and actually knowing it inside-out. That is, in my opinion, the reason why the humble oldie one is so damn reliable and accurate in many respects. It has the genius of the original design and programmimg. The new one, carry the flaws of understanding of posterior engineers who probably never grasped the original's subtleties. The new one also had to deal with further limitations as it tries to do much, much more (oscilloscope) with a more powerful, but yet still limited uC. They probably had to shrink down the code of the component tester to fit everything in this design.
Absolutely. I only took a look at the capacitance measurement method and it's amazing. I seriously doubt that FNIRSI has done the same due diligence in developing the TC3 and the outcome is obvious. Thanks for your comment!
Hello. My transistor tests lcr. (DIY-M12864) which is an older variant of the yellow lcr meter you have. uses a chip. atmega328p-pu (20mhz) (8bit) similar to the one used in your lcr meter but mine is a long older style package and not square.
If it's not bad enough, the reading also depends on which set of slots you use. Try the same component in different sets of "1,2,3" slots and you'll get a different reading with this one. It's all over the place and unreliable. Maybe you got 2 completely different readings is because you used a different set of slots, not because you wrote it down wrong.
On the lcr-t4 device we see a platform for testing the necessary SMD elements in Sot-23, TO-252 and other cases. This completes the comparison of the necessary device with the toy from finirsi. Someone may object that this platform can be made independently, because the manufacturer of the toy did not understand why it is needed and did not even include it in the kit. I will say that I have been using the lcr-t4 device for many years, I use new custom firmware that is freely available.
I'm sure the bad quality ZIF sockets have their share in the ESR reading. Would have been interesting to see a comparison to the values with the cap soldered to the pads.
Yes, I appreciate that leads affect the ESR - sometimes I got wildly off readings and wiggling the cap in the ZIF would fix that. But after all this is what the tool is offering. I tend to test SMD caps by pressing them hard on the pads of the old tester after adding a little solder on the legs so there is a small lump of solder protruding.
I would typically go with 100Hz or 120Hz on the ESR meter when testing caps that are going to be used in devices working with rectified AC line voltages or transformed variations thereof. When working on switching power supplies, class D amplifiers, RF or digital circuits, I'll step up the test frequency to as high as the ESR meter will go, 10kHz or 100kHz if available, because you really don't want capacitors that act like resistors in those kind of devices. Some service manuals even make mention of the need to replace caps in class D amps with low ESR caps when making substitutions.
@@Tony359_2that's not strictly correct, because especially on older DC/DC converter designs you can get nesty oscillations due of the regulation loop by using low ESR components. This is due to the design of it. It was considered to begin with. And there are other situations as well. Lower is not strictly better. Some designs relay on a certain ESR range. I did such a design many years back with a product series running for many year. Out of the sudden problems with newly batches came up. It turned out someone on the supply side substituted a EOL cap with a fancy new one...
Thanks! Yes, of course sometimes you are out of luck and need to replace a like for like. On older devices, anything modern should be ok but that's also not always the case as I'll discover in one of my next videos where I re-cap an old CRT monitor :)
I think, before comparing these two testers and or the datasheet information, you should have started your testing using a quality Capacitor Tester as the base line compared to the datasheet. Then you would have a more accurate component reading to compare to. Just my thoughts.... From the cheap seats..!
@Tony359_2 And you still wonder why it shows up as diodes after putting it straight from one tester to the other, as shown in the video. A tester uses current to test and change the capacitor.
The original documentation mentions that if a charge is detected (a small one) the cap is discharged first. And my old tester never fails to identify a capacitor. I do discharge large capacitors before testing them, for small ones the tester should be able to take care of that. Anyways, the "2 diodes" issue is really not the issue here :)
I built a ESR meter many years ago. It was a nightmare to calibrate. It was very good when you got it calibrated but because of the sensitivity it was a pain to setup . I use the same open tester as it is a go nogo . But I have a commercial capacitance tester that i can use for more accuracy. The problem with that is it doesn't measure very small capacities . As i work on RF circuits alot it is difficult to measure them and a commercial tester for them is way out of my price range.
For designing purposes it must be difficult to get a good tool at a reasonable price indeed! I was thinking of building an ESR meter myself for checking in circuit, there is a project from an italian youtuber, Pier Aisa I think. I see it's being used by many creators!
@@Tony359_2 I wish you good if you build it . Things have both easier and harder what I mean by that with the hobbyist access to SMD a lot better layouts can be used, the bad part is it's SMD. My hands are just not suited to the fine work needed.
Try long pressing, or double pressing, the button after it has found no component (and before it resets to try testing again). Should bring up a menu, including option to calibrate, on most transistor testers. Unless they've literally removed all that in their custom build of firmware... ATMega can easily output a 100khz square wave, but not sine. You'd need filtering for that. So probably four dedicated pins, with filters, and opamp buffers, to produce the most common sine test signals. Not really in the spirit any more of just using the 6 resistor connections to make a versatile meter. The main IC has probably been chosen to be able to do the oscilliscope & colour LCD function correctly, that is more much compute challenging for most tasks than what the code for the TransistorTester does. I'd guess some clone-STM32 chip.
I had that thought but didn't find an STM32-like or a Chinese RiscV with a matching pinout so far. It looks like a potential match for an msp430 though.
Thanks for trying but Is your method clean? Some times you start with the old tester, sometimes the new one, sometimes you discharge, mostly you don’t?
Thanks for your comment - discharging small caps should not matter. I tried a few times to see if it changed anything. the devices can discharge small caps - bigger ones must be discharged or the IC will blow - ask me how I know it. Regarding the method... point taken but this is my second channel where things are less polished :)
Testing very low ESR would the resistance in the component leads and socket connector result in a higher reading? As for the DE-5000, I have one. That device uses a 4-wire measurement to compensate for the length of the leads. It requires an open/short calibration at the desired frequency if you want the highest accuracy.
It makes sense - to be fair those tools are not meant to return a perfect reading but yes, they should be both calibrated - the old one cannot be calibrated unfortunately. Interesting about the 4 wires, I was wondering how the DE-5000 was dealing with those probes. Thanks!
I think the changes in the values might be because of bad contact in the zif socket, i have one of the older meters and have soldered wires and added clips at the end to grab on the components, and the values became more consistent, those sockes tend to oxidate and become loose with use, more so being of low quality
Indeed - That said, Fnirsi has just released a component tester (only). Will that be up to the old design? Or, in other words, will have managed to make a better copy this time? :)
Tony you might want to mention this in a video.... The best way to help with "the algorithm" is to click like on as many of the channel maker's comments as you can, so look for the channel LOGO under "replies" and find Tony's [here] and like it. You can disregard any other comments as it makes less of a difference to youtube. ;8^) TH-cam seems to like this activity.
Thanks. I honestly don't think anybody knows how the algorithm works. There are many speculations but there is no evidence for anything. You're very welcome to like my comments, then I'll let YT do their stuff, I cannot possibly change that a lot :)
The old tester is calibrated with 3 wires connect between them and in the process requires a capacitor with a capacitance greater than 100 nF. The 3 wires are placed in the first 3 holes on the top side from the left.
This is my playlist about this tester: th-cam.com/play/PLXLg86Nx7KCypJ3LIee-1Zc6QaAxMvBvc.html&si=RmHohQwxsClb8cv3
19 วันที่ผ่านมา
fnirsi makes great tools but they're firmwares are a bit wonky. They've sent me the DWS-200 soldering iron for a review but the temperature was so off that I had to tell them about that, looks like they're working on a fix tho, maybe this component tester just needs better firmware?
Thankfully I didn’t get that station!! The specs mention a linear supply but then I saw a review where they showed an SMPS one. I guess it’s no more FNIRSI for me - though the USB tester works fine!!
If I'd have to hazard a blind guess, I'd assume they mashed together circuitry from the component tester and the "DSO" on the measuring inputs without ever for a moment considering how one might affect the other - so what was artfully bodged into working fairly well on the original tester might be getting dragged off kilter by whatever else is on those pins in the FNIRSI...
@@Tony359_2 though it has one big improvement - mine also has a decent sized color screen. But I wouldn't be surprised, if the magic inside is identical clone.
The fact that the TC3 could not identify the resistance in series with the capacitor correctly is obviously a huge red flag, their firmware isn't even sure if it's a capacitor or 'anti-parallel' diodes half the time. ESR = Equivalent Series Resistance, all you did was add a series resistance, only thing that *might* throw things off is inductance. Film resistors are much better about not having much any inductance. But that's a fail for the TC3 otherwise. Also your yellow board SHOULD let you calibrate it by shorting all 3 together then hitting the button, but always possible that function was simply ripped out for whatever reason.
I have tried many methods and pushed that button, hold it, sing while pushing... it never works so I have to assume it was removed for whatever reason indeed.
@@Tony359_2 Ah classic rookie mistake! Have to stand in a bucket of water on one leg, rotate clockwise while holding unit in your left hand, press the button each 1/16th of rotation precisely and chant, It's just that easy. I have a totally identical unit fyi (except probably in firmware loaded) v2.07 shows when it boots, and I was able to run through calibration via the expected method, which I know I've never even tried until now.
ahh, the LEFT hand! I am right-handed, wrong hand then. I guess it that's why it doesn't work with me. I'll find someone left-handed so we can calibrate it! ;)
I've never found these devices to be accurate. They're useful little tools and essential to have in the toolkit due to the range of stuff they can test - but take readings with a large pinch of salt. I've had issues with spurious ESR readings specifically - I have a proper ESR tester (a Peak Atlas ESR60) and the measurements never seem to match up all that well. There is, however a £30, unbranded tester available from the usual places that 12voltvids reviewed recently that uses a tweezers setup for its probes. This unit matches up with what the Peak says pretty well and has the added benefit of being able to measure at multiple frequencies.
Btw, when testing adding a resistor to a cap you forget to discharge the cap after and you did mention they left a charge on them and meter did not like it.
Thanks for another nice video. I'm curious about a quick comparison when you would be able to get your "proper" meter. A nice second part to this one, perhaps ??? Hello @Fluke !!!! Hello @Peak !!!! You there ????
I have a number of PEAK testers, they have been really useful, quick to use and reliable and I use them a lot. They aren't particularly expensive, given what you get for your money (£60-£150 or so depending on what you get). They do a pack of the LCR bridge and Digital component tester, plus a case for less than £200. Being specific to particular jobs means that you pick and choose the tools you need. For example the Zener tester has a higher voltage source compared to the rest of the units as it can test up to 50V Zener diodes. The higher spec digital component tester DCA75 lets you do things like curve tracing on transistors via and app on the PC. They have jigs for things like SOT23 transistor testing. These capabilities are not present on the generic testers. I've had no problems with reliability whatsoever. Obviously you do have to do things like making sure that capacitors are discharged before testing the, but that's not unique to this product either.
@@timc3600 I could not agree more. I am very happy with the atlas esr70 gold (my first product from Peak) and am thinking of buying more from them when I need them. As Tony said, these products are for life and well worth the investment.
I'm pretty sure i got into calibration mode on the LCR T4 somehow. I think it involved a lot of button holding and waiting, to get past the display calibration screen and then get to measurement calibration one. But ultimately i'm not convinced i can recommend it, the factory setup is perfectly fine. It's not precision calibration anyway. It mostly relies on component values anyway and the calibration i think it just cancels out the PCB capacitance or something like that. Yo the disguised microcontroller, so the casing ejector pattern, the area cleared by laser, and the pinout (oscillator connected down the middle of the first side) looks like it might be an MSP430 of some persuasion.
Hello Tony! Can I ask you, why did you use resistor in the phase of calibration, shoud you not have to use someting near zero reistance, like a copper paperclip?
Hello - I didn't use a resistor. I used three "resistor legs" - I always keep them as sometimes you need a piece of solid wire to make a jumper :) Thanks for watching!
Function Generator 1,5 khz offset ( only square wave correct), component tests often suspect , only the Scope funktion 100% reliable....but for the money - these times, there a better scopes on the Market !
wellll.... when I tested the scope on the main channel, at first it read 8V and after running the AUTO mode it switched to 12V. 12V was the correct one. No, there are much better ones for just a little more on the market IMHO. Thanks for watching!
....A CALIBRATION bug could easily make for bizarre behavior. ...I would be interested in seeing if repeated 'calibration' makes the TC3 give different results.... i.e. calibrate and test same component 10 times and see if results vary wildly. ...With luck you may find one particular calibration gives sensible results. ...I can easily imagine a knock off device might be reasonably accurate when it leaves the factory with the default settings but goes horribly wrong as soon as user 'calibrates' it.
The readings are even worse than you think. The ESR ratings in the data sheets are maximum, after the capacitor has been in circuit for it's rated life (100hrs, 5000hrs, etc) at the maximum rated temperature. They should be even lower when brand new.
@@Tony359_2 If you want to be scientific yes the $300 LCR meter is right one but if you want to measure thing with not much higher accuracy the TC-1 in my opinion is the right one. Ther is even an advanced one from PEAK LCR-45.
Well, I'd assume Peak would be happy to repair it for you. That's how it normally works! :) Yes, I probably don't need an LCR meter but the DE-5000 is £150 and maybe it's a "nice tool to have" instead of buying cheapo ones and then wondering if the reading is good or not :) But thanks anyways!
Hello. I have DIY-M12864 diy kit with white blue box. with calibration function. to enter the menu you must hold down the power/select button for a few seconds. if you buy the cheapest ones, you cannot expect high accuracy! I have 3 different cheap lcr meters and you will get 3 different results. even if the results are not so different, they are not the same.
At least the original tester does discharge the capacitor (up to some level) before testing. Anyways the result never changed on the TC3 on all my tests.
Hey Tony, your old "generic" tester IIRC has its ZIF worn out. Keeping that in mind it might be plausible that contacts resistance is around 0.3ohms. If that's the case you got no chance to measure ESR less than 0.3ohms in the best case scenario. On the other side, don't throw away FNirsi tester. FNirsi is a more or less "known brand" company so it's possible that they will come up with an firmware update that would solve the issue you're having and would make this tool back into a "usable thingy" status.
Yes it could be. I tried some contact cleaner on it though. But anyways as I said I don't expect that level of accuracy. 0.x Ohm is totally fine. For more accurate readings, I need a better tool! I won't throw it away (thanks!) but I'm tired of these useless tools! :)
@@Tony359_2 heh, that's the cost of the popularity. The more popular you get - the more useless tools you'd be sent to deal with :-D. Not the worst thing ever TBH.
ahahah - to be honest I look forward when proper brands will approach me with decent tools to review :) OR - I'm ok with smaller brands but the tool has to work fine! :) Oh well, I am a dreamer.
@@Tony359_2 The man is allowed to dream if you ask me. And yet again, keeping in mind the quality of the content you produce it is just a matter of time till you'd be in the top tier of retro-tech-tubers. If you ask me - and I'm following your main channel for quite a while - you're already up there with Adrian, Chris "gadget UK", CRG, Jan Beta to name a few, but it would take some time for major sponsors to start nagging you with fine offers.
There will be no refund, he was sent it for free to review it.... but let's just say, it's unlikely Finirsi will sell many of these in its current design.
Correct, I forgot to mention it on this video but it was a free unit. I think I might invest in a new LCR meter - and maybe fix the ZIF socket indeed!!
@@Tony359_2 I found the sockets can also be disassembled, cleaned, terminals reformed straight, and the lever lubed. A little tricky to get back together. If successful, saves soldering wear on your motherboard. However, if the plating has worn through, cleaning will only be temporary.
Those things you hold the components in those testers are crap.. depending on how far you put the leads in or dirt on the contacting surfaces changes. I removed mine and soldered in a proper screw terminal on it and dont get as much fluctuation..
I'm not in Australia, I am free to say what I want! Yes, I did get in touch, they replied with some useless stuff so they are aware. And no, there is no FW update at this stage.
well... you can see it the other way round, proper equipment is not cheap to develop and manufacture. The DE5000 is not too expensive. My view is that my older tester is a good surrogate if a proper LCR meter is not available. The TC3 is not.
@@Tony359_2 You are right totally, but from a hobby POV this would't make any sense. I also at the end, replace all the caps of my projects anyways, but it sucks is not more accurate, I was actually looking into getting one Also there's a new FNIRSI DWS 200 solder station, it'd be great if you can take a look at it. After watching your channel I am just contemplating getting a jbc.. honestly I have had many problem with cheap instruments, I think it's time to step it up your work is greatly appreciated
@@nR-kv7xo The DWS has a switching power supply and as such I'd steer clear from it - see my Aixun T3A review on the main channel th-cam.com/video/B7jLpHgR7fI/w-d-xo.html (unless you want to fry your ICs!) The JBC is too expensive - but as you say there's a time where you have to decide how much time you want to spend on your tools...
This matches my experience with this FNIRSI tester. Some of their products are pretty ok for their price, but this is not one of them. Just like you, I bought one, several months ago, hoping to replace my old generic transistor tester with this, and the readings were a huge disappointment. Mostly with capacitors. I compared several caps against my generic transistor tester, a DER LCR meter, a handmade ESR meter, and two different LCR tweezers. Turns out the generic tester was a lot closer to actual usable readings. Got rid of the FNIRSI immediately. The one good thing about the fnirsi is its higher voltage testing for zener diodes, and the fact that it's also a basic oscilloscope for learning or audio/slow signals. But as a component tester it's not great.
@@Tony359_2 Hey it's not like Fnirsi did a lot better on their 50€ tool. They just took the same shitty TFXTDOL sockets and threw them into a pot of Rit black fabric dye, and then covered the branding and fake patent number with the front panel.
I laugh every time yet another serious engineer reviews a new piece of s̶h̶ "measuring" equipment from this so called brand and talks about how totally worth its money a device is or mentions "question marks" regarding its performance, while it's crystal clear for anyone with minimal eyesight that it is a complete and utter garbage.
ahahah - I know. Unfortunately Fluke have not been in touch yet! :) I had the USB tester from FNIRSI (I bought it myself) and it is a solid piece of tool. Not this one. I have rejected further offers from FNIRSI from now. Some of those small tools can be good/decent. Just take the original (yellow) tester: it really does work and there is quite some work behind it. But too many copycats around.
It is almost a slander to say an ATMega is basically an Arduino. It should be the other way around. Also, it is capable of frequency outputs of 4 MHz when driven at 8 MHz.
@@Tony359_2 Would be fun to see what could be done with the firmware in these testers. If the hardware connections are well thought out the firmware may be much improved.
Sul mio tester cinese ho un menu con una funzione di self test e calibrazione che prima mi chiede di mettere in corto i tre pin del socket, poi mi chiede un condensatore mi pare da 20 picofarad. Magari è più recente del tuo... Comunque per adesso il FNIRSI costa troppo. 😂
Lascialo pure dov'e'! :) La calibrazione era nel tester originale quindi non e' chiaro perche' il mio non l'abbia. Ma sono cinesate. copie di copie di copie. Chissa' dove si sono persi la sezione della calibrazione :D
![เขามัทรี - เอ็กซ์ ศุภกฤต - จอนนี่มิวสิค [ Official MV ]](http://i.ytimg.com/vi/AS9ufKuQIVs/mqdefault.jpg)
Quite ironic that FNIRSI went to the length of lasering off the IC markings for a device no one will want to replicate!
Ahah 😂
It's pitiful that any company would do such a thing. I've got a Cosori air fryer which I'd love to fix but the markings on the chip that has failed have also been lasered away!
definitely an STM something or a customized FPGA of a Z80 ;o)
I have the same yellow component tester since 2017. I wanted to have as you said, something to identify a component as good/bad and if good, how good it is (ESR, hfe, etc). After using it for many years, I get to trust it .
Last year, for my birthday, I got another component tester, TC1. The main differences are the color display, the plastic case and the rechargeable battery. As functionality, the TC1 seems to give me nore details about transistors, but its measurements are in par with the old one, maybe some 5-10% difference. However, now I can not swear that a resistor is 220 or 226 Ohm.
What I say is that you should trust the one that give you the most consistent values when test a component; I see that your DSO-TC3 gives different values at each testing, that hints to a error in firmware or in the schematic or the comonents they used.
You mean the DSO-TC3 in the video?
I've been using the yellow one for quite some time and as you say, I learnt to trust it, I never felt it was giving me inaccurate results. Of course, as I said, I've always been looking for "good" "bad" outcome, nothing more.
I had bought this meter from an estate sale for pennies and was noticing that it was giving off readings as well. I'm glad these two videos came up on your channels Tony.
Fnirsi random number generator. Might be useful for picking your lottery numbers. 😅
ahaha - WINNING numbers?
Thanks Tony, I use the 100Hz frequency as I find it results in a clearer result between good and bad caps, higher frequencies result in a narrower spread in ESR between good and bad caps, making it less obvious what their condition is like.
No, thank you Scott!
Good to know about the frequency, I hope I can experiment myself soon!
Would have been good to do extra tests at 1khz/10khz to see how much the ESR changed. The TransistorTesters original firmware claims to produce a value more accurate to a 10khz test
@@ivololit was still nowhere near it, I had done that off camera.
So, I'm not surprised that voltage measurement would not see the frequency, as you are measuring the capacitor voltage, and smoothing the voltage is the role of the capacitor.
What I'm thinking is measure the current going to the capacitor on the oscilloscope (like voltage drop of a 1ohm resistor in series). I assume you could see current going 2 ways with the measurement frequency.
Those ATMega tester don't work at high frequencies, it's all on the document linked in the description. Anyways, I doubt an electrolytic cap can do much at 100KHz.
The FNIRSI TC3 is a multi function device being primarily an oscilloscope. The component test function is secondary. I use the component test as a go/no go indication and to find out the value of random capacitors. For that it works very well. If I were interested in ESR and other measurements, I would buy a more expensive device with switchable frequencies. You need a high frequency for low pfs and a low frequency for electrolytics.
I think maybe you expect too much from the TC3 which, for the price, and it isn't a dedicated component tester, actually works very well. Not many of us have three different devices to compare test results!
I think it's fair to expect when they present a device as "that cheap LCR of the kind you used before, but now with a scope and extra features and much more expensive", that the scope and all those other features might be bad, but at least the LCR functionality doesn't work worse than you're used to! It would have been my assumption for sure.
indeed. I am comparing the TC3 to a £5 device, not to a Fluke LCR meter :) I thought it would match that!
Primarily an oscilloscope? Really? :)
There is no primary/secondary, there is just an "overall bad".
This device aims to be a "I can do everything" device. But as usual: Something able to do everything, does everything more or less bad.
I wouldn't recommend either, not the scope function nor the component tester. And in the end, this device got compared to a dirt cheap component tester, but failed horrible. Yeah, meh...
I think they tried to port the old AVR code to some kind of ARM device, with completely different I/O port characteristics, but failed to modify the code in a proper way. The original includes a lot of constants and mathematical tricks to get to the point. I assume they tried something, they don't even understand on a basic level. So, it might be fixable with firmware updates, but... as Chinese manufacturer usually handle this, I won't expect any improvement here.
@@matthiash.4670 absolutely. And I understand they have just released a new "component tester ONLY" device just a few days ago.
When the best part of the tool is its case :D
ahahaha it is, it looks very cool! :D
Thanks, saved me the money for a bad product (Finirsi)
Their software regarding functionality is an absolute disaster/garbage, its made to look good, NOT to work properly...
I have 2 of those kind of component testers. They work ok for semiconductors, but for caps, resistors and inductors, I trust my dedicated LCR meters way more. If I want more detailed swept frequency measurements of LCR’s, I use my Analog Discovery with its impedance-measurement attachment. When I worked in a development lab, there I ahd access to very high-ended LCR-bridges, that are unaffordable to a normal mortal on retirement, but the Analog discovery comes close, except if you want a DC bias. Those cheap component testers are great for simple semiconductor tests, but for more detailed tests, curve tracers are the way to go. With all that being said, wish those component testers were available when I started in electronics in the 1970’s.
You are absolutely right, those testers are great for what they do but because they say "ESR" on it it won't mean it's going to be accurate.
It's good from a hobbyist perspective as you say as long as we are all aware of the limitations!
Hi , Tony, impedance is the sum of Xc (1/2*Pi*f*C) , esr and esl ( 2*Pi*f*L) .Thats what the datasheet shows at 100 kHz.😊
Thank you, I realised that a bit too late - does that mean that impedance = ESR at 100KHZ?
@@Tony359_2 It is the sum of the reactance of the capacitance plus the Equivalent Series Resistance plus Equivalent Series Inductive Reactance.Xc + Resr + Xl.If the total impedance at is for example 15 mOhm , you should subtract Xc and Xl to get Resr.The only correct method is a LCR bridge meter who can extract ESR and capacitance correctly at a certain frequency.DER-EE DE-5000 LCR meter is a good precise meter at a reasonable price.
Thank you much for the follow up. ❤
The new meter seems to be a combination of the DSO-138 "Oszilloskop" and the old component tester. But as chinese standard: a bad copy without proper (german) engineering.
I have the old tester and now i know to keep it, because the new thing isnt worth any pennys.
It's so disappointing indeed! :)
@@Tony359_2 For sure! A bit more investigation how and why the germany guy has made his software as it is would come up in a very nice "multitool". But this thing is useless.
It is possible to put "original" firmware into the old-type tester. Don't know for M-Tester, but for original firmware, it is recommended to measure a capacitor across terminals 1 and 3.
thanks - A user got in touch to point out I might be able to upgrade my old tester. Thanks for your input as well!
There's no hard reason for that i don't think. Other than layout symmetry reasons and tiny little capacitive differences. On the schematic and microcontroller all 3 pins are identical.
Between 1-3 was worse, I showed this in my live stream.
@@TheDefpom Original FW has a special RC mode for capacitors plugged between 1 and 3.
Ty for the test super info sir :)
You are very welcome :)
And markings have been removed. I bet they use at least some Open Source components for their firmware (if not most of it) and then they have the audacity to do that. Not that it is very difficult to find out what the chip is anyway. BTW. Most of Open Source licenses require to publish the source code (not the binaries) together with any modifications. I wonder if they do that. And I think the original component tester of this kind was entirely Open.
Oh I'm certain they didn't develop a new method for that £30 tester :) Yes, the original tester is fully documented online.
For entering selftesting/calibrating mode on the classic tester you should (according to the original documentation) short pins 1, 2,3, then press the button and press it again within 2 seconds. If this doesn’t work, the function was disabled when compiling the firmware.
Thanks. I just tried - nothing. The screen blinks (while testing) but nothing happens.
Great video Tony, Thank You. I've been wondering about the FNIRSI Products, as I'm re-populateing my workspace after being burgled recently.
So, Hyper Tinkering.! Sounds great 😎😆.
I don't know why, but today is the first time yt has let come to this channel. There is No Subscribe button, and No Notification button.??
Oh no, I am so sorry to hear that. I hope you get back up and running very soon.
I don't know, you should see a subscribe button somewhere. There shouldn't be anything special about this channel, it's just an independent one, I just own both.
@Tony359_2
I'm not sure what changed, but I am Subscribed now 👍😎😆✅
I prefer to build my own gear and equipment, drag all those years of math and theory into the real world. (lol)
Maybe this is a hint to update / upgrade.
I count myself lucky, that I didn't have any client jobs at the time.
I have an separate esr meter and ive watched it with osc. Can confirm it can test at 10khz and even 100khz. Other bonus is that it has leads and in most cases desoldering isn't needed for caps, it doesn't trigger any semiconductors
If you don't want to test other components and just want a somewhat affordable and perhaps slightly more reliable LCR meter, maybe check out the JYE Tech M162. It's a DIY kit for around 50 Euro, and specializes in just LCR (no semiconductors). But you can manually select what you want to test (no auto-id that goes awry in edge cases). And it'll not only tell you capacitance, but also ESR, impedance, Q and D value, and theta. Testing frequency is selectable between 100 Hz and 1 KHz, so it does use actual frequency testing.
JFTR: It gave plausible results during my experiments, but I have not done any extensive, proper testing with expensive, calibrated equipment. So basically: caveat emptor.
thank for that, it looks like a good option. I'm expecting an LCR meter from a small manufacturer - it's apparently a decent one. I could not source the DE-5000, it's impossible to find it here in the UK and the manufacturer could not help. Thanks again!
Basically, the impedance is NOT the same as ESR. Also, the math to calculate the values from assumed equivalent circuit to another - assumed series connection on the ESR or assumed parallel connection of the leakage (!) resistor - involves quite challenging list of formulas. General Radio used to print the collection in their LCR meter manuals. I think I still have one of those manuals, maybe something like GR 1607 or 1608 somewhere. You might find them through Google search, I have not tried. Anyway, I suspect the ATMega or other chip can do conversions (calculations), running with 8 MHz or other high speed clock, while it is probably not trying to make the actual primary measurements at any of the customary frequencies.
Thanks and sorry for assuming Impedance = ESR :)
The ATmega is not using any high frequency at all, it's all described in the manual linked in the description but there is no HF involved, it's a very clever approach :)
I'd buy two meters more to divide my certainty by two. Or a cheap oscilloscope with an square wave generator (rectified sine or pure sine would also do) to be certain of how exact the values are. With tons of ´patience, a bridge.
Go for it Tony. :)
Or maybe not :)
Excellent testing,, Tony, especially when you go back to the original design to understand how it is measuring capacitance. Maybe a firmware update for the DSO-TC3 will help, but most people will be interested in this unit as a component tester, not a signal generator or oscilloscope. It really needs to be as good if not much better than then the original component tester it is clearly designed to replace.
Indeed, I've had my tester for many years, I thought a more modern replacement could at least match it but no.
Someone has mentioned the TC3 is primarily an oscilloscope... I don't understand! :D
It seems that FNIRSI is well-aware of the flaws of this product. Evidently, they've released the new LCR-P1 component tester in the past couple of days to fill this role suitably. There is a thread on EEVBlog forums, but it's recent enough that nobody has it delivered yet.
Also, judging from the fact that the listing cleverly refuses to use the term LCR anywhere except the SKU identifier, I'd still be doubtful of them redeeming themselves even on second try.
ahahah amazing! What's the point in advertising the TC3 then?
Not only do you need the test frequency, but you need to be able to change the frequency. Different caps under test require different freqs. You wouldn't test a 3300uF cap at 100KHz and you wouldn't test a 33pF cap at 120Hz. I got a DE-5000 and haven't looked back. I only use these testers for IDing mystery transistors/diodes.
For sure. Though after many years I never felt that old tester ever fooled me. Not accurate by any mean - the original designer confirms that - but it gives you a good idea.
I'm trying to get a DE-5000, not easy in the UK apparently thanks to some arguable political decisions with our borders...
@@Tony359_2 Hello, Tony! Great video as always! Now, I've never had LCR meter, so I decided to splurge some money on my New Year gift and got Peak Atlas ESR70 Gold. It's made in UK and I got it because many repair shops recommend it to me and online reviews were also pretty favorable. De ree's was out of option because of the price. Maybe you could check Peak's one out?
I am indeed - though I like the DE-5000's ability to select frequencies manually and down to 100Hz.
Apart from my 1st comment on the possible reasons why the Fnirsi lacks in comparison to the original, and now trying to answer you on the question of the "how" and "what frequency" a capacitor tester operates...
Well, capacitor testers usually are designed around two different approaches.
One is the "DC" approach or let's call it, the "Charge - Discharge" approach.
The other is the "AC approach" or "reactance" approach.
Proper LCR meters use the second method and actually apply an frequency to the capacitor (or inductor) and measure its impedance (the complex sum of the resistive and reactive reactances of the component).
The mini-testers, however, use the DC approach so... No, they're not using any frequency so to speak.
They short the capacitor (to establish a known zero charge starting point) and then they apply a known voltage with a known resistance in series, and use the capacitor charging formula to estimate the capacitor value based on the time it takes for it to charge from zero up to a certain intermediate voltage.
This is usually what multimeters with capacitance measurement do...
The AVR based tester, being micro-controlled, can do a few more "clever" things, like measuring the voltage at different points in time and, by comparing the actual charge discharge curve to the theoretical exponential curve, it can derive some of the capacitor imperfections.
For the ESR, for instance, it first makes an estimation of the capacitance value based on a first "time-to-charge" method... And after it has a baseline, it can then measure it again using, for instance, a different charging resistor, expecting to see a calculated change in the time-to-charge.
But when it actually charges in a different time than the calculated, it can infer the ESR and the loss.
The method has limited capacities because the small device don't have that many different resistors and charging speeds to obtain an ideal measurement of a large range of capacitors so, it's probably better at measuring those things for a certain range of capacitor values, while for other values, it probably looses resolution amd precison but still can calculate a rounded off value.
It's quite a clever method.
it's an incredibly clever method indeed! Plenty accurate for my needs :)
Hi Tony, datasheets usually list the worst-case conditions. About the measurement of 5 ohms with the resistor in series, in my opinion, maybe because that's its limit. Maybe by design and because it would be a damaged component, it doesn't measure more than 5 ohms.
Anyway, it's useful.
On the main channel video it read 16.3 Ohms on a bad cap - the other would read 110Ohm :)
Yes, it's very useful. A great paperweight :)
@@Tony359_2 LOL Maybe OK should be read as KO :)
Despite the appearance, the old and rather amateur looking one is actually an exact implementation of the original design by the german guy who had the original idea.
Even though it had its limitations, the designer did quite an extensive job in perfecting it and correcting for its shortcomings.
Cut many years past, to the FNIRSI product, the concept (of the original design) was quite known already and replicated and modified in subsequent implementations to a point that many people and many companies felt "like" having "mastered" the original concept but...
As with anything involving low level programming, it goes a great distance between one "thinking" he understood what a program does and one actually understanding it and actually knowing it inside-out.
That is, in my opinion, the reason why the humble oldie one is so damn reliable and accurate in many respects.
It has the genius of the original design and programmimg.
The new one, carry the flaws of understanding of posterior engineers who probably never grasped the original's subtleties.
The new one also had to deal with further limitations as it tries to do much, much more (oscilloscope) with a more powerful, but yet still limited uC.
They probably had to shrink down the code of the component tester to fit everything in this design.
Absolutely. I only took a look at the capacitance measurement method and it's amazing. I seriously doubt that FNIRSI has done the same due diligence in developing the TC3 and the outcome is obvious. Thanks for your comment!
Hello. My transistor tests lcr.
(DIY-M12864)
which is an older variant of the yellow lcr meter you have.
uses a chip.
atmega328p-pu (20mhz) (8bit) similar to the one used in your lcr meter but mine is a long older style package and not square.
to the ESR of the Panasonic Co the contact resistance (you can see at "calibrating try" on the old one) is added - as you soldered the 33Ohms.
If it's not bad enough, the reading also depends on which set of slots you use. Try the same component in different sets of "1,2,3" slots and you'll get a different reading with this one. It's all over the place and unreliable. Maybe you got 2 completely different readings is because you used a different set of slots, not because you wrote it down wrong.
The tester that keeps surprising you :) Amazing thanks!
On the lcr-t4 device we see a platform for testing the necessary SMD elements in Sot-23, TO-252 and other cases. This completes the comparison of the necessary device with the toy from finirsi. Someone may object that this platform can be made independently, because the manufacturer of the toy did not understand why it is needed and did not even include it in the kit. I will say that I have been using the lcr-t4 device for many years, I use new custom firmware that is freely available.
Thanks, I'll see if I can update mine too!
I'm sure the bad quality ZIF sockets have their share in the ESR reading. Would have been interesting to see a comparison to the values with the cap soldered to the pads.
Yes, I appreciate that leads affect the ESR - sometimes I got wildly off readings and wiggling the cap in the ZIF would fix that. But after all this is what the tool is offering. I tend to test SMD caps by pressing them hard on the pads of the old tester after adding a little solder on the legs so there is a small lump of solder protruding.
Thanks. Good info I can use.
You're welcome!
Nice Thinking. Informative.
Thanks, appreciated!
I would typically go with 100Hz or 120Hz on the ESR meter when testing caps that are going to be used in devices working with rectified AC line voltages or transformed variations thereof. When working on switching power supplies, class D amplifiers, RF or digital circuits, I'll step up the test frequency to as high as the ESR meter will go, 10kHz or 100kHz if available, because you really don't want capacitors that act like resistors in those kind of devices. Some service manuals even make mention of the need to replace caps in class D amps with low ESR caps when making substitutions.
I tend to buy low ESR caps when I get some so I know I can use them everywhere. Thanks for watching!
@@Tony359_2that's not strictly correct, because especially on older DC/DC converter designs you can get nesty oscillations due of the regulation loop by using low ESR components. This is due to the design of it. It was considered to begin with. And there are other situations as well.
Lower is not strictly better. Some designs relay on a certain ESR range.
I did such a design many years back with a product series running for many year. Out of the sudden problems with newly batches came up. It turned out someone on the supply side substituted a EOL cap with a fancy new one...
Thanks! Yes, of course sometimes you are out of luck and need to replace a like for like. On older devices, anything modern should be ok but that's also not always the case as I'll discover in one of my next videos where I re-cap an old CRT monitor :)
@@Tony359_2 glad you came across this effect also! Will definitely watch ur upcoming video. Thx for teasing it!
I think, before comparing these two testers and or the datasheet information, you should have started your testing using a quality Capacitor Tester as the base line compared to the datasheet. Then you would have a more accurate component reading to compare to. Just my thoughts.... From the cheap seats..!
You are totally right - unfortunately I don't have that and that's why I mentioned Defpom's video in the video :)
I think the fault is in your test actions. You should deploy a capacitor to empty for a long enough time before placing it in a tester each time.
As long as the capacitor is not dangerously charged, the tester will take care of that. And I showed I discharged it as a test in the video.
@Tony359_2 And you still wonder why it shows up as diodes after putting it straight from one tester to the other, as shown in the video. A tester uses current to test and change the capacitor.
The original documentation mentions that if a charge is detected (a small one) the cap is discharged first. And my old tester never fails to identify a capacitor.
I do discharge large capacitors before testing them, for small ones the tester should be able to take care of that. Anyways, the "2 diodes" issue is really not the issue here :)
I built a ESR meter many years ago. It was a nightmare to calibrate. It was very good when you got it calibrated but because of the sensitivity it was a pain to setup . I use the same open tester as it is a go nogo . But I have a commercial capacitance tester that i can use for more accuracy. The problem with that is it doesn't measure very small capacities .
As i work on RF circuits alot it is difficult to measure them and a commercial tester for them is way out of my price range.
For designing purposes it must be difficult to get a good tool at a reasonable price indeed! I was thinking of building an ESR meter myself for checking in circuit, there is a project from an italian youtuber, Pier Aisa I think. I see it's being used by many creators!
@@Tony359_2 I wish you good if you build it . Things have both easier and harder what I mean by that with the hobbyist access to SMD a lot better layouts can be used, the bad part is it's SMD. My hands are just not suited to the fine work needed.
My issue with SMD is storage. It's like storing sand.
Try long pressing, or double pressing, the button after it has found no component (and before it resets to try testing again). Should bring up a menu, including option to calibrate, on most transistor testers. Unless they've literally removed all that in their custom build of firmware...
ATMega can easily output a 100khz square wave, but not sine. You'd need filtering for that. So probably four dedicated pins, with filters, and opamp buffers, to produce the most common sine test signals. Not really in the spirit any more of just using the 6 resistor connections to make a versatile meter.
The main IC has probably been chosen to be able to do the oscilliscope & colour LCD function correctly, that is more much compute challenging for most tasks than what the code for the TransistorTester does. I'd guess some clone-STM32 chip.
I'll try but I spent a few minutes doing exactly that the other day :)
I had that thought but didn't find an STM32-like or a Chinese RiscV with a matching pinout so far. It looks like a potential match for an msp430 though.
Capacitors are normally tested between 1-3 pin for accuracy
Between 1-3 was WORSE ! I showed the different pin combinations when testing this in my live stream.
Thanks for trying but Is your method clean? Some times you start with the old tester, sometimes the new one, sometimes you discharge, mostly you don’t?
Thanks for your comment - discharging small caps should not matter. I tried a few times to see if it changed anything. the devices can discharge small caps - bigger ones must be discharged or the IC will blow - ask me how I know it.
Regarding the method... point taken but this is my second channel where things are less polished :)
Testing very low ESR would the resistance in the component leads and socket connector result in a higher reading? As for the DE-5000, I have one. That device uses a 4-wire measurement to compensate for the length of the leads. It requires an open/short calibration at the desired frequency if you want the highest accuracy.
It makes sense - to be fair those tools are not meant to return a perfect reading but yes, they should be both calibrated - the old one cannot be calibrated unfortunately.
Interesting about the 4 wires, I was wondering how the DE-5000 was dealing with those probes. Thanks!
I think the changes in the values might be because of bad contact in the zif socket, i have one of the older meters and have soldered wires and added clips at the end to grab on the components, and the values became more consistent, those sockes tend to oxidate and become loose with use, more so being of low quality
if you're talking about the TC3, it's brand new :)
Too bad, it would have been nice with a more modern and better replacement for the good old component tester!
Indeed - That said, Fnirsi has just released a component tester (only). Will that be up to the old design?
Or, in other words, will have managed to make a better copy this time? :)
cheap tester more accurate if take contact resistance into account?
Tony you might want to mention this in a video.... The best way to help with "the algorithm" is to click like on as many of the channel maker's comments as you can, so look for the channel LOGO under "replies" and find Tony's [here] and like it. You can disregard any other comments as it makes less of a difference to youtube. ;8^) TH-cam seems to like this activity.
Thanks. I honestly don't think anybody knows how the algorithm works. There are many speculations but there is no evidence for anything. You're very welcome to like my comments, then I'll let YT do their stuff, I cannot possibly change that a lot :)
The old tester is calibrated with 3 wires connect between them and in the process requires a capacitor with a capacitance greater than 100 nF. The 3 wires are placed in the first 3 holes on the top side from the left.
unfortunately shorting the top row doesn't make a difference.
I found this vídeo: th-cam.com/video/O7NwkdKEUdU/w-d-xo.html
@@Tony359_2 I think this tester is LCR T4
This is my playlist about this tester: th-cam.com/play/PLXLg86Nx7KCypJ3LIee-1Zc6QaAxMvBvc.html&si=RmHohQwxsClb8cv3
fnirsi makes great tools but they're firmwares are a bit wonky. They've sent me the DWS-200 soldering iron for a review but the temperature was so off that I had to tell them about that, looks like they're working on a fix tho, maybe this component tester just needs better firmware?
Thankfully I didn’t get that station!! The specs mention a linear supply but then I saw a review where they showed an SMPS one. I guess it’s no more FNIRSI for me - though the USB tester works fine!!
@@Tony359_2 It's no more finirsi for you anway, after your review they are definitely not sending you any more free stuff hehe!
You'd be surprised, even Aixun wanted to send me their new station!
FNIRSI also offers firmware updates for download from their site - current version there is 0.30 maybe that improves things?
I did check that of course and the TC3 was on 0.30 since the beginning unfortunately but thanks
Maybe the "new" one is capped at 5 ohm? :)
No, it read 16 ohm when I tested a bad cap on the main channel's video
If I'd have to hazard a blind guess, I'd assume they mashed together circuitry from the component tester and the "DSO" on the measuring inputs without ever for a moment considering how one might affect the other - so what was artfully bodged into working fairly well on the original tester might be getting dragged off kilter by whatever else is on those pins in the FNIRSI...
you might be right :)
Try holding the button down when you turn it on for the calibration menu (for the old one)
I did but it didn't work unfortunately. Thanks for suggesting.
I have a cheap 15€ component tester from China with model name of TC1. Works beautifully.
It's probably similar to mine and likely 100% based on the original german one! :) Mine also works ok!
@@Tony359_2 though it has one big improvement - mine also has a decent sized color screen. But I wouldn't be surprised, if the magic inside is identical clone.
excellent
thanks!
The fact that the TC3 could not identify the resistance in series with the capacitor correctly is obviously a huge red flag, their firmware isn't even sure if it's a capacitor or 'anti-parallel' diodes half the time.
ESR = Equivalent Series Resistance, all you did was add a series resistance, only thing that *might* throw things off is inductance. Film resistors are much better about not having much any inductance. But that's a fail for the TC3 otherwise. Also your yellow board SHOULD let you calibrate it by shorting all 3 together then hitting the button, but always possible that function was simply ripped out for whatever reason.
I have tried many methods and pushed that button, hold it, sing while pushing... it never works so I have to assume it was removed for whatever reason indeed.
@@Tony359_2 Ah classic rookie mistake! Have to stand in a bucket of water on one leg, rotate clockwise while holding unit in your left hand, press the button each 1/16th of rotation precisely and chant, It's just that easy.
I have a totally identical unit fyi (except probably in firmware loaded) v2.07 shows when it boots, and I was able to run through calibration via the expected method, which I know I've never even tried until now.
ahh, the LEFT hand! I am right-handed, wrong hand then. I guess it that's why it doesn't work with me. I'll find someone left-handed so we can calibrate it! ;)
I've never found these devices to be accurate. They're useful little tools and essential to have in the toolkit due to the range of stuff they can test - but take readings with a large pinch of salt.
I've had issues with spurious ESR readings specifically - I have a proper ESR tester (a Peak Atlas ESR60) and the measurements never seem to match up all that well. There is, however a £30, unbranded tester available from the usual places that 12voltvids reviewed recently that uses a tweezers setup for its probes. This unit matches up with what the Peak says pretty well and has the added benefit of being able to measure at multiple frequencies.
Ah interesting! I'll take a look thanks!
Btw, when testing adding a resistor to a cap you forget to discharge the cap after and you did mention they left a charge on them and meter did not like it.
Well, I blew one of those testers with a motor start capacitor charged to 220V!! 5V charge is ok-ish on them. I think :)
Thanks for another nice video. I'm curious about a quick comparison when you would be able to get your "proper" meter. A nice second part to this one, perhaps ??? Hello @Fluke !!!! Hello @Peak !!!! You there ????
I hope Fluke replies LOL! :)
I have a number of PEAK testers, they have been really useful, quick to use and reliable and I use them a lot.
They aren't particularly expensive, given what you get for your money (£60-£150 or so depending on what you get). They do a pack of the LCR bridge and Digital component tester, plus a case for less than £200. Being specific to particular jobs means that you pick and choose the tools you need. For example the Zener tester has a higher voltage source compared to the rest of the units as it can test up to 50V Zener diodes. The higher spec digital component tester DCA75 lets you do things like curve tracing on transistors via and app on the PC. They have jigs for things like SOT23 transistor testing. These capabilities are not present on the generic testers.
I've had no problems with reliability whatsoever. Obviously you do have to do things like making sure that capacitors are discharged before testing the, but that's not unique to this product either.
@@timc3600 I could not agree more. I am very happy with the atlas esr70 gold (my first product from Peak) and am thinking of buying more from them when I need them. As Tony said, these products are for life and well worth the investment.
definitely considering them - the DE-5000 is not easy to purchase here in the UK
@@Tony359_2 If you are close to LON, I could borrow mine 🙂
I'm pretty sure i got into calibration mode on the LCR T4 somehow. I think it involved a lot of button holding and waiting, to get past the display calibration screen and then get to measurement calibration one.
But ultimately i'm not convinced i can recommend it, the factory setup is perfectly fine. It's not precision calibration anyway. It mostly relies on component values anyway and the calibration i think it just cancels out the PCB capacitance or something like that.
Yo the disguised microcontroller, so the casing ejector pattern, the area cleared by laser, and the pinout (oscillator connected down the middle of the first side) looks like it might be an MSP430 of some persuasion.
Does the FNIRSI just max out at reading 5ohms ESR ? I wonder how it would react if you used a lower value resistor, under 5ohms?
No, it read 16 Ohm on the other video on the main channel.
Hello Tony!
Can I ask you, why did you use resistor in the phase of calibration, shoud you not have to use someting near zero reistance, like a copper paperclip?
Hello - I didn't use a resistor. I used three "resistor legs" - I always keep them as sometimes you need a piece of solid wire to make a jumper :) Thanks for watching!
If you only looking to see if component is good or bad, perhaps look at Mr Carlson lab he has made long time ago a capacity tester.
I might have watched that video some time ago
As incredible as it may seem, the tc1 can do what this one can't... a disappointment.
Exactly!
Function Generator 1,5 khz offset ( only square wave correct), component tests often suspect , only the Scope funktion 100% reliable....but for the money - these times, there a better scopes on the Market !
wellll.... when I tested the scope on the main channel, at first it read 8V and after running the AUTO mode it switched to 12V. 12V was the correct one. No, there are much better ones for just a little more on the market IMHO.
Thanks for watching!
You can buy a (EAST TESTER ET430) for less then $150 and it works great
it looks interesting, thanks!
To calibrate the old unit short the first 3 pins on the top row
I tried that too. Top row is a bit mangled, that’s why I didn’t use it in the video.
Older tester seems to have consistent offset. I wonder if it could be callibrated with a programmer.
I understand the FW can be updated! :)
....A CALIBRATION bug could easily make for bizarre behavior.
...I would be interested in seeing if repeated 'calibration' makes the TC3 give different results.... i.e. calibrate and test same component 10 times and see if results vary wildly.
...With luck you may find one particular calibration gives sensible results.
...I can easily imagine a knock off device might be reasonably accurate when it leaves the factory with the default settings but goes horribly wrong as soon as user 'calibrates' it.
Maybe! But I am not willing to invest so much time on that device, I'll wait for FNIRSI to fix the bug :)
The readings are even worse than you think. The ESR ratings in the data sheets are maximum, after the capacitor has been in circuit for it's rated life (100hrs, 5000hrs, etc) at the maximum rated temperature. They should be even lower when brand new.
Good point, datasheet are always the worst scenario of course. Thanks.
You have to buy the TC-1.. it's more accurate sometimes and you can trust it
Yes, others have said that! I think if I buy something it's going to be a proper LCR meter :) Thanks!
@@Tony359_2 If you want to be scientific yes the $300 LCR meter is right one but if you want to measure thing with not much higher accuracy the TC-1 in my opinion is the right one.
Ther is even an advanced one from PEAK LCR-45.
Well, I'd assume Peak would be happy to repair it for you. That's how it normally works! :)
Yes, I probably don't need an LCR meter but the DE-5000 is £150 and maybe it's a "nice tool to have" instead of buying cheapo ones and then wondering if the reading is good or not :)
But thanks anyways!
Hello. I have
DIY-M12864 diy kit with white blue box.
with calibration function.
to enter the menu you must hold down the power/select button for a few seconds.
if you buy the cheapest ones, you cannot expect high accuracy!
I have 3 different cheap lcr meters and you will get 3 different results. even if the results are not so different, they are not the same.
I tried holding the button, nothing happens :)
10:50 Maybe discharge before switching between testers ?
At least the original tester does discharge the capacitor (up to some level) before testing. Anyways the result never changed on the TC3 on all my tests.
Hey Tony, your old "generic" tester IIRC has its ZIF worn out. Keeping that in mind it might be plausible that contacts resistance is around 0.3ohms. If that's the case you got no chance to measure ESR less than 0.3ohms in the best case scenario. On the other side, don't throw away FNirsi tester. FNirsi is a more or less "known brand" company so it's possible that they will come up with an firmware update that would solve the issue you're having and would make this tool back into a "usable thingy" status.
Yes it could be. I tried some contact cleaner on it though. But anyways as I said I don't expect that level of accuracy. 0.x Ohm is totally fine. For more accurate readings, I need a better tool!
I won't throw it away (thanks!) but I'm tired of these useless tools! :)
@@Tony359_2 heh, that's the cost of the popularity. The more popular you get - the more useless tools you'd be sent to deal with :-D. Not the worst thing ever TBH.
ahahah - to be honest I look forward when proper brands will approach me with decent tools to review :)
OR - I'm ok with smaller brands but the tool has to work fine! :) Oh well, I am a dreamer.
@@Tony359_2 The man is allowed to dream if you ask me. And yet again, keeping in mind the quality of the content you produce it is just a matter of time till you'd be in the top tier of retro-tech-tubers. If you ask me - and I'm following your main channel for quite a while - you're already up there with Adrian, Chris "gadget UK", CRG, Jan Beta to name a few, but it would take some time for major sponsors to start nagging you with fine offers.
That's very very kind of you to put me on the same level of some of my favourite creators, thank you so much!
Those 14p connectors are easy to find and cheap, too. Probably best to send back the FNIRSI unit and use the refund for the new blue connector.
There will be no refund, he was sent it for free to review it.... but let's just say, it's unlikely Finirsi will sell many of these in its current design.
Correct, I forgot to mention it on this video but it was a free unit. I think I might invest in a new LCR meter - and maybe fix the ZIF socket indeed!!
You can get ARIES sockets if you don't want the cheap teal knockoffs nor want to pay 3M pharmacy prices, something in between.
@@Tony359_2 I found the sockets can also be disassembled, cleaned, terminals reformed straight, and the lever lubed. A little tricky to get back together. If successful, saves soldering wear on your motherboard. However, if the plating has worn through, cleaning will only be temporary.
I disassembled it some time ago and I cannot remember why I could not improve it. But yes, I'd rather invest the time replacing it :) Thanks though!
measure a known bad cap in both!
I did that! :)
Those things you hold the components in those testers are crap.. depending on how far you put the leads in or dirt on the contacting surfaces changes. I removed mine and soldered in a proper screw terminal on it and dont get as much fluctuation..
Absolutely but that's what Fnirsi supply :)
Ooooo the FNIRSI lawyers will be after you for this one 😂 curious if there is a firmware update available or if you tried to contact them?
I'm not in Australia, I am free to say what I want! Yes, I did get in touch, they replied with some useless stuff so they are aware. And no, there is no FW update at this stage.
I hope I didn't sound too harsh towards my Australian friends - and I hope that "stunt" backfires massively, it's such a silly thing to do.
@g4z-kb7ct
Well, actually...
Wow Original Best...
Is your fnirsi fully charged ?
I think it is? You can see the battery level in the video.
This really sucks. Proper equipment are way too expensive.
well... you can see it the other way round, proper equipment is not cheap to develop and manufacture. The DE5000 is not too expensive. My view is that my older tester is a good surrogate if a proper LCR meter is not available. The TC3 is not.
@@Tony359_2 You are right totally, but from a hobby POV this would't make any sense.
I also at the end, replace all the caps of my projects anyways, but it sucks is not more accurate, I was actually looking into getting one
Also there's a new FNIRSI DWS 200 solder station, it'd be great if you can take a look at it.
After watching your channel I am just contemplating getting a jbc.. honestly I have had many problem with cheap instruments, I think it's time to step it up
your work is greatly appreciated
@@nR-kv7xo The DWS has a switching power supply and as such I'd steer clear from it - see my Aixun T3A review on the main channel th-cam.com/video/B7jLpHgR7fI/w-d-xo.html (unless you want to fry your ICs!)
The JBC is too expensive - but as you say there's a time where you have to decide how much time you want to spend on your tools...
This matches my experience with this FNIRSI tester. Some of their products are pretty ok for their price, but this is not one of them.
Just like you, I bought one, several months ago, hoping to replace my old generic transistor tester with this, and the readings were a huge disappointment. Mostly with capacitors. I compared several caps against my generic transistor tester, a DER LCR meter, a handmade ESR meter, and two different LCR tweezers. Turns out the generic tester was a lot closer to actual usable readings. Got rid of the FNIRSI immediately.
The one good thing about the fnirsi is its higher voltage testing for zener diodes, and the fact that it's also a basic oscilloscope for learning or audio/slow signals. But as a component tester it's not great.
indeed - the USB tester I purchased was good so I kind of assumed other tools were of good quality. But not all as you say!
You can open the old socket and tighten it.
I tried it and I couldn’t do that.
I have socket trouble as well after 8 years. Some of the plastic spars broke. Better get a fresh socket. They aren't exactly using premium ones.
you wouldn't expect anything premium on a £5 tool :)
@@Tony359_2 Hey it's not like Fnirsi did a lot better on their 50€ tool. They just took the same shitty TFXTDOL sockets and threw them into a pot of Rit black fabric dye, and then covered the branding and fake patent number with the front panel.
With what I earn from the videos I can barely cover the electricity costs for making the videos! :D
I laugh every time yet another serious engineer reviews a new piece of s̶h̶ "measuring" equipment from this so called brand and talks about how totally worth its money a device is or mentions "question marks" regarding its performance, while it's crystal clear for anyone with minimal eyesight that it is a complete and utter garbage.
ahahah - I know. Unfortunately Fluke have not been in touch yet! :)
I had the USB tester from FNIRSI (I bought it myself) and it is a solid piece of tool. Not this one. I have rejected further offers from FNIRSI from now.
Some of those small tools can be good/decent. Just take the original (yellow) tester: it really does work and there is quite some work behind it. But too many copycats around.
It is almost a slander to say an ATMega is basically an Arduino. It should be the other way around.
Also, it is capable of frequency outputs of 4 MHz when driven at 8 MHz.
Absolutely, my knowledge on the subject is limited! Thanks!
@@Tony359_2 Would be fun to see what could be done with the firmware in these testers. If the hardware connections are well thought out the firmware may be much improved.
Sul mio tester cinese ho un menu con una funzione di self test e calibrazione che prima mi chiede di mettere in corto i tre pin del socket, poi mi chiede un condensatore mi pare da 20 picofarad.
Magari è più recente del tuo...
Comunque per adesso il FNIRSI costa troppo. 😂
Lascialo pure dov'e'! :)
La calibrazione era nel tester originale quindi non e' chiaro perche' il mio non l'abbia. Ma sono cinesate. copie di copie di copie. Chissa' dove si sono persi la sezione della calibrazione :D
@@Tony359_2 beh il mio ha anche un rotary encoder, sul tuo il menu proprio non c'è. Magari qualcuno sul FNIRSI ci fa un firmware custom...
eh, con l'integrato grattato diventa difficile da farsi. Ma chissa'.
magari è un STM32.
Probabilmente lo e', non avrebbe senso "reinventare la ruota"