Just to let you know Shahriar, I have this device, E4980A in my Lotto Lab dream wish list, if I win the lotto I will have one on my bench for sure. Thank you for sharing with me the fantastic job you have done for this man who loves test gear.
There are MUCH BETTER ones available from Keysight! Cost enough to put a strain on your lotto winnings! Even this one is over 20 grand to fully outfit it!
Some aircraft use capacitive fuel level sensors by measuring the difference between air and fuel. It’s quite sensitive and very robust. That made a very good demonstration of the operating principle.
Yes , some theory with hands are best Shahriar , I whish you hade made more like this video on some simple superheterodyne radio receiver , IQ modulation , FM modulation , PLL, PLO , spectrum analyzer basic system block ( which is acutely is superheterodyne receiver ) I personally I think one of best video you make it was with you brother ,in that with video you two made best approach learning method . Regards from Sweden
Paul, check out Shahriar's spectrum analyzer and signal generator repair videos. They aren't just about the repair itself---In several of these he goes through some of the topics you mention. They are well worth the watch, even if you aren't interested in the particular instrument or the specific repair he is performing.
Thank you so much for all the videos you poste. You are the best teacher. Thank you again for taking the effort of making those videos and sharing the knowledge with us.
That was very interesting and informative. I'd never even considered electrolytics resonating at such a low frequency! The bias feature of that unit is great and adds an additional dimension to the devices that can be tested. I was planning to make a 137MHz tube based NOAA satellite receiver which requires a 2-15pF mechanical variable capacitor. I think I'll use some diodes instead! Thanks as always. btw, I've noticed that TH-cam has been deleting my comments on your channel. God knows why. TH-cam makes me facepalm almost every day. Don't know whether you'll see this comment.
@@Thesignalpath Thanks for the response, appreciated :) Another creator that I exchange email with outside of TH-cam emailed me last week to say that he has noticed that many comments are deleted on his channel before he gets to respond to them. He emailed me directly in response to one of my comments on his channel to tell me that my comment was gone before he got a chance to respond. I'm sure the same is happening everywhere. Let's just call it the TH-cam AI dystopia. Looking forward to your next review/teardown. Cheers.
Those Keysight 4980A(s) are around $20k new! I just bought a couple at work and one showed up with a screw loose inside, needless to say I was a little disappointed, for such an expensive device! I could have popped it open and figured out where it came from but, I sent in said meter, and it took around 4 weeks to get back. But it was calibrated again so that may have contributed to the time.
very nice experiments. I started getting "obsest" by caps about 10 years ago trying to measure fF. I designed and build a meter and started collecting bridges (like the GR1520 that was used in GRs cal lab. The funny thing is that I learned about the fringing effect from building my own caps using PCB. But I used only one piece of double sided pcb and calculated the capacitance.
A good test of the inconsistencies across the surface would be to place the skewer in the middle as you have already done and each side and multiple places in between.
Since you have this amazing instruments, can you make one video on OPEN and SHORT corrections, how to perform them and in which situations they can be useful. No one touches this aspect but I think it is quite important and interesting to see the effect of it.
Great video! Thx. I was lucky to buy 4263B for a good price only downside is frequencies and low dc bias. Mainly for electrolytics is useful to have bias up to rated voltage.
For the electrolytic & tantalum, using a 1V test signal, shouldn't a DC bias be applied so that the cap doesn't get reverse biased on the negative half cycle?
Please note that the actual voltage swing present across the device is only a fraction of the applied voltage. You can see the voltage on the device on the screen (as well as the current through the device). But you are right that under the circumstances where the frequency is low and the capacitance is also low, the user should keep an eye on the voltage across the DUT to make sure it does not reverse bias the capacitor strongly.
@@Thesignalpath Thanks. A very interesting demonstration would be to show how some of the high-K ceramic dielectrics have a very strong voltage dependence (Z5U, Y5V) where the capacitance decreases pretty dramatically vs. DC bias. While dielectrics such as X7R are much better - of course, not as good as NP0/C0G.
@@Thesignalpath Shahriar, I saw it at 12:40. At the risk of replying to an old video, could you elaborate on why this is the case that the voltage swing present across the device is only a fraction of the applied voltage? Are parasitics the cause of such a big difference? I imagine it can't be, otherwise it would render the measurement useless.
It would be interesting to see the equivalent parallel resistance of a bunch of equal valued capacitors of different dielectrics. Particularly with film caps. Polystyrene is legendary for low leakage, but I recall an article by Bob Pease about polypropylene's superiority in this regard.
Since you have this marvelous instrument (hopefully with option 001) could you measure a 1 to 10 uF MLCC (6V, 20V, 50V) while sweeping the DC Bias to ~ 40 V (inst. max)? This would be quite enlightening to many engineers who have these capacitors in buck, or especially boost power supplies of >12V as to why there is so much ripple and instability which is not shown in the SPICE models?
A lot of MLCC caps are piezoelectric, and expand when a voltage is applied over them. This in turn reduces their capacitance and in turn they are worse at suppressing ripples at the higher voltages. Were designing a voltage multiplier for a Penning gauge, and the caps I looked at dropped in capacitance by over 30% when at their max voltage (700 or so volts) compared to their capacitance with a 0 volt charge.
Are there any tips on frequency range, bias voltage and type of fixtures to be used to measure capacitance as small as ~ 50fF? Should one measure the parasitic capacitance first? Also, what things to be kept in mind while measuring such small capacitance?
When I was at a transmitter station in the Royal Navy the transmitters had variable vacuum capacitors in them and the piece of test gear used to test them was so dangerous to use, only senior rates could do the test while us mere mortals watched. It was basically a very high voltage power supply with a current meter, I can't remember if it was nano, micro or milliamps, but I remember that with the failed capacitor the current slowly started to rise as the voltage was turned up which was in excess of 10kV.
My guess is that the internal block level architecture contain a high resolution DAC driving the test voltage/frequency and a high res ADC after current ranging transimpedance amplifier. My questions are: 1. Why not plot Z/Y over frequeny? 2. Why not show HD for a component, e.g. ceramic caps? 3. Why not fit to a 3 or more equivalent network? 4. Why not place at least a DAC buffer and the transimpedance amp on the DUT testboard?
Hi, I saw your video and it is very useful. I have a handheld LCR and trying to fully utilize its functionality. In your video about measuring 1000uF electrolytic capacitor at 100 KHz, its capacitance value is no longer meaningful. However I have a question about the ESR value at 100KHz for this 1000uF capacitor? in your video, the LCR meter is still displaying something, is this still an accurate ESR value for this capacitor at 100 KHz?
Sorry, must be my OLD ears playing tricks on me. Nevertheless, I suggest my point is still pertinent. My 'life' is in the kilo and mega arena while your life considers that to be barely above DC!
Hi, I'm Eric. I want to use 2 LCR meters (Keysight E4980A) to measure the inductance changes of several inductive coils, and the input signal has to be in the same phase due to the coil design needs. Is it possible to achieve this kind of thought? and also, do you know how to do it?
Is it possible to achieve a degradation mechanism for an electrolytic capacitors using an LCR meter by either considering a thermal or voltage overstress. I am aware one can adjust the voltage for the DUT of the LCR meter but is it possible to have the capacitors placed in an oven and measure with the LCR meter continuously. I look forward to your response.
maybe I understoot it wrong but I think Vx at the high pot is a voltage meter, not a generator. Or did you mean there is the guarding source ( but not visible in the drawing)that puts a voltage on the coax shield ? There is a lot of info on the site of IET, they scanned all the GR docs they could find with the help of bridge guru Henry Hall ( very nice person who worked 40 years for GR as designer. He gave me very good advise when I was restoring my GR 1608. It is a few GB of info and very well and comprehensive. It is from the time engineers learned how things really worked 😎 ( like you ) watched all your videos, they really are a class on its own. I made some videos about more advanced measuring and I know how much work this is. I stopped because my internet is to slow, it took more then a day to upload and often needed to do it several times because internet dropped. So thank you.
I think it would be intresting to actually show the difference between auto balancing and not, and the sometimes surprising situations where it matters.
Curious, Applent Anbai LCR meters like AT810A or AT3818 has similar physical instrument layout and LCD menu. Anbai seems manufacturing LCR meters for GW Instek and perhaps they are also manufacting for Keysight?
I think Agilent was doing this design before Applent existed. Given the HP /Agilent/Keysight history of high end test equipment I think there is a good chance Applent designed theirs to resemble the earlier design.
Could you answer me something? How precision companies know how much is 1 volt? How they precisely get that information "from" so they can precisely calibrate multimeters and so on?
It's important to know that Volt is not a fundamental unit. The SI base unit is kg*m^2/(A*s^3). There are calibration standards for those units. More practically, there are voltage standards based on various technologies that can accurately produce a set voltage (see wikipedia for Josephson voltage standard for example). From there, there is the concept of traceability and NIST certificates and so on, so that you can transfer the standard from one device to another.
If you want some search terms, google "electrical metrology" which is the field encompassing this, and "Josephson junction" which is the modern standard of a volt. These devices are maintained by NIST in the US and other orgs worldwide, that maintain a traceable chain of standards that is used to calibrate devices.
@@supernumex May I add that the traceability does not guarantee the accuracy of any unit, it does only guarantee that if a unit used to test your unit, is found faulty, is it possible to warn you about the error. Every test gear used from the highest level and to user level is within a margin of error that gets transferred to the unit tested by it. One more thing is that most units do not get calibrated as in adjusted in accuracy but is tested for deviation and change in accuracy from new to now. Often is it more important to know how your unit changes over time than try to bring it back.
It comes from very accurately instruments (transfer standards), which in turn are calibrated using some sort of primary standard at a standards lab, for instance the national metrology institute of the respective country. For voltage, that primary standard could e.g. be based on a Josephson junction: en.wikipedia.org/wiki/Josephson_voltage_standard
Hi Shahriar. Such a great review. Would you be able to compare this to a much cheaper instrument like a DER EE DE-5000 for us would cannot immediately afford a the better E4980A.
Then if you find one you see they REALLY want to hold onto them when you see the price! You might have better luck finding a 4275, which is a 1980's LCR meter that goes to 10 MHZ. It is a LOT larger physically (the 19" rack box that is typical of HP instruments of that era) than these new instruments but it has comparable performance. I got mine in working condition off ebay for $600.
You may want to consider an Advantest R8340A as an alternative. It combines a HV DC source and a low current meter to make a very high resistance meter. They are much cheaper on ebay and still have good performance. I have one and have had good luck with it.
Fantastic instrument if you have close to 20 grand to buy one! I am amazed at how you manage to get literally millions of dollars worth of instruments! Super rich patrons? You won the lotto? Sold your soul? :)
Thank you so much! It's so nice to see those things you've been taught, and have just accepted for years, actually demonstrated. Great, great job.
Shahriar…. Great video. Thanks for posting!
I really enjoy your experiment videos! I hope you make more of them
Just to let you know Shahriar, I have this device, E4980A in my Lotto Lab dream wish list, if I win the lotto I will have one on my bench for sure. Thank you for sharing with me the fantastic job you have done for this man who loves test gear.
There are MUCH BETTER ones available from Keysight! Cost enough to put a strain on your lotto winnings! Even this one is over 20 grand to fully outfit it!
I enjoyed the lecturelike style of this video a lot. Thank you!
Glad you enjoyed it!
Some aircraft use capacitive fuel level sensors by measuring the difference between air and fuel. It’s quite sensitive and very robust. That made a very good demonstration of the operating principle.
Great video. I learned a lot. Thank you!
I want to see a tear-down of that!
dont turn it on, take it apart......oh wrong channel...
Yes , some theory with hands are best Shahriar , I whish you hade made more like this video on some simple superheterodyne radio receiver , IQ modulation , FM modulation , PLL, PLO , spectrum analyzer basic system block ( which is acutely is superheterodyne receiver )
I personally I think one of best video you make it was with you brother ,in that with video you two made best approach learning method . Regards from Sweden
I will try to make more basic videos too. The next extensive tutorial is being planned out.
Paul, check out Shahriar's spectrum analyzer and signal generator repair videos. They aren't just about the repair itself---In several of these he goes through some of the topics you mention. They are well worth the watch, even if you aren't interested in the particular instrument or the specific repair he is performing.
Thank you so much for all the videos you poste. You are the best teacher. Thank you again for taking the effort of making those videos and sharing the knowledge with us.
That was very interesting and informative. I'd never even considered electrolytics resonating at such a low frequency! The bias feature of that unit is great and adds an additional dimension to the devices that can be tested. I was planning to make a 137MHz tube based NOAA satellite receiver which requires a 2-15pF mechanical variable capacitor. I think I'll use some diodes instead! Thanks as always.
btw, I've noticed that TH-cam has been deleting my comments on your channel. God knows why. TH-cam makes me facepalm almost every day. Don't know whether you'll see this comment.
Glad you liked it! I hope this comment stays up. I replied to it so you know that I read it. :)
@@Thesignalpath Thanks for the response, appreciated :) Another creator that I exchange email with outside of TH-cam emailed me last week to say that he has noticed that many comments are deleted on his channel before he gets to respond to them. He emailed me directly in response to one of my comments on his channel to tell me that my comment was gone before he got a chance to respond. I'm sure the same is happening everywhere. Let's just call it the TH-cam AI dystopia. Looking forward to your next review/teardown. Cheers.
Loved the demonstration, fantastic!
Waiting for the day you decide to brighten up the world with some diy build projects on measurement devices. So much junk out there.. I vote yes! 😄
Electrical engineers: "Jmaginary"
Yeah those math guys that pretend that i = sqrt(-1) ... go figure, when we all know it is j. ;)
Iurrent was already taken
@@ikocheratcr it equals both!
Those Keysight 4980A(s) are around $20k new! I just bought a couple at work and one showed up with a screw loose inside, needless to say I was a little disappointed, for such an expensive device! I could have popped it open and figured out where it came from but, I sent in said meter, and it took around 4 weeks to get back. But it was calibrated again so that may have contributed to the time.
That’s pretty insane lol, quality missed it. I wonder if they do vibration testing on these units to pick up anything
Great tutorial!!!
neat experiment. i will have to try it out for myself :)
very nice experiments. I started getting "obsest" by caps about 10 years ago trying to measure fF. I designed and build a meter and started collecting bridges (like the GR1520 that was used in GRs cal lab. The funny thing is that I learned about the fringing effect from building my own caps using PCB. But I used only one piece of double sided pcb and calculated the capacitance.
A good test of the inconsistencies across the surface would be to place the skewer in the middle as you have already done and each side and multiple places in between.
Since you have this amazing instruments, can you make one video on OPEN and SHORT corrections, how to perform them and in which situations they can be useful. No one touches this aspect but I think it is quite important and interesting to see the effect of it.
Great video! Thx. I was lucky to buy 4263B for a good price only downside is frequencies and low dc bias. Mainly for electrolytics is useful to have bias up to rated voltage.
For the electrolytic & tantalum, using a 1V test signal, shouldn't a DC bias be applied so that the cap doesn't get reverse biased on the negative half cycle?
Please note that the actual voltage swing present across the device is only a fraction of the applied voltage. You can see the voltage on the device on the screen (as well as the current through the device). But you are right that under the circumstances where the frequency is low and the capacitance is also low, the user should keep an eye on the voltage across the DUT to make sure it does not reverse bias the capacitor strongly.
@@Thesignalpath Thanks. A very interesting demonstration would be to show how some of the high-K ceramic dielectrics have a very strong voltage dependence (Z5U, Y5V) where the capacitance decreases pretty dramatically vs. DC bias. While dielectrics such as X7R are much better - of course, not as good as NP0/C0G.
@@Thesignalpath Shahriar, I saw it at 12:40. At the risk of replying to an old video, could you elaborate on why this is the case that the voltage swing present across the device is only a fraction of the applied voltage? Are parasitics the cause of such a big difference? I imagine it can't be, otherwise it would render the measurement useless.
Do you have videos using that equipment in transformers and core's (toroide)???? 👋
It would be interesting to see the equivalent parallel resistance of a bunch of equal valued capacitors of different dielectrics. Particularly with film caps. Polystyrene is legendary for low leakage, but I recall an article by Bob Pease about polypropylene's superiority in this regard.
Since you have this marvelous instrument (hopefully with option 001) could you measure a 1 to 10 uF MLCC (6V, 20V, 50V) while sweeping the DC Bias to ~ 40 V (inst. max)? This would be quite enlightening to many engineers who have these capacitors in buck, or especially boost power supplies of >12V as to why there is so much ripple and instability which is not shown in the SPICE models?
Sure, it is definitely possible. The instrument has all the options.
@@Thesignalpath 😊 Hint: the capacitance will be < than 20% at the rated voltage.
@@shazam6274 Be sure you are using the frequency the manufacturer used to define the capacitance of the device.
@@glasslinger This is not about frequency (which is shown in the capacitor's SRF graph). It is about value change with DC bias
A lot of MLCC caps are piezoelectric, and expand when a voltage is applied over them. This in turn reduces their capacitance and in turn they are worse at suppressing ripples at the higher voltages.
Were designing a voltage multiplier for a Penning gauge, and the caps I looked at dropped in capacitance by over 30% when at their max voltage (700 or so volts) compared to their capacitance with a 0 volt charge.
Super interesting, thank you.
Are there any tips on frequency range, bias voltage and type of fixtures to be used to measure capacitance as small as ~ 50fF? Should one measure the parasitic capacitance first? Also, what things to be kept in mind while measuring such small capacitance?
What about a teardown of this instrument?
thank you
Do the VNA work the same way when measuring capacitance at low-ish frequency?
No, a VNA still measure incident and reflected waves. From that of course you can compute the impedance.
When I was at a transmitter station in the Royal Navy the transmitters had variable vacuum capacitors in them and the piece of test gear used to test them was so dangerous to use, only senior rates could do the test while us mere mortals watched.
It was basically a very high voltage power supply with a current meter, I can't remember if it was nano, micro or milliamps, but I remember that with the failed capacitor the current slowly started to rise as the voltage was turned up which was in excess of 10kV.
My guess is that the internal block level architecture contain a high resolution DAC driving the test voltage/frequency and a high res ADC after current ranging transimpedance amplifier. My questions are:
1. Why not plot Z/Y over frequeny?
2. Why not show HD for a component, e.g. ceramic caps?
3. Why not fit to a 3 or more equivalent network?
4. Why not place at least a DAC buffer and the transimpedance amp on the DUT testboard?
Hi, I saw your video and it is very useful. I have a handheld LCR and trying to fully utilize its functionality. In your video about measuring 1000uF electrolytic capacitor at 100 KHz, its capacitance value is no longer meaningful. However I have a question about the ESR value at 100KHz for this 1000uF capacitor? in your video, the LCR meter is still displaying something, is this still an accurate ESR value for this capacitor at 100 KHz?
... "Or even in the mere terahertz frequencies"....
Huge sigh... I live in an ENTIRELY different world to you sir...
Pretty positive he said "near terahertz"... but still, yeah.
near-THz, definitely not mere-THz! :)
Sorry, must be my OLD ears playing tricks on me.
Nevertheless, I suggest my point is still pertinent.
My 'life' is in the kilo and mega arena while your life considers that to be barely above DC!
why the instrument is showing Cp values while there is no DUT connected?
Hi, I'm Eric. I want to use 2 LCR meters (Keysight E4980A) to measure the inductance changes of several inductive coils, and the input signal has to be in the same phase due to the coil design needs. Is it possible to achieve this kind of thought? and also, do you know how to do it?
The best models for femto farads and Atto farads measurement at high frequency.
Is it possible to achieve a degradation mechanism for an electrolytic capacitors using an LCR meter by either considering a thermal or voltage overstress. I am aware one can adjust the voltage for the DUT of the LCR meter but is it possible to have the capacitors placed in an oven and measure with the LCR meter continuously. I look forward to your response.
maybe I understoot it wrong but I think Vx at the high pot is a voltage meter, not a generator. Or did you mean there is the guarding source ( but not visible in the drawing)that puts a voltage on the coax shield ? There is a lot of info on the site of IET, they scanned all the GR docs they could find with the help of bridge guru Henry Hall ( very nice person who worked 40 years for GR as designer. He gave me very good advise when I was restoring my GR 1608. It is a few GB of info and very well and comprehensive. It is from the time engineers learned how things really worked 😎 ( like you )
watched all your videos, they really are a class on its own. I made some videos about more advanced measuring and I know how much work this is. I stopped because my internet is to slow, it took more then a day to upload and often needed to do it several times because internet dropped. So thank you.
What is the "D" value shown on the display?
D denotes dissipation factor.
I think it would be intresting to actually show the difference between auto balancing and not, and the sometimes surprising situations where it matters.
The auto balancing cannot be turned off unfortunately.
Curious, Applent Anbai LCR meters like AT810A or AT3818 has similar physical instrument layout and LCD menu. Anbai seems manufacturing LCR meters for GW Instek and perhaps they are also manufacting for Keysight?
I think Agilent was doing this design before Applent existed. Given the HP /Agilent/Keysight history of high end test equipment I think there is a good chance Applent designed theirs to resemble the earlier design.
These instruments had a Japanese origin I think. The are the successors of the older HP branded ones.
can I ask you a question please, would a hi frequency lcr meter (5 Mhz) be of any use in a pcb repair lab?
already have a hp 4263A up to 100KHz.
For "repair", maybe not so much. But for design, sure.
@@Thesignalpath thank you so much for replaying.
Could you answer me something?
How precision companies know how much is 1 volt?
How they precisely get that information "from" so they can precisely calibrate multimeters and so on?
It's important to know that Volt is not a fundamental unit. The SI base unit is kg*m^2/(A*s^3). There are calibration standards for those units. More practically, there are voltage standards based on various technologies that can accurately produce a set voltage (see wikipedia for Josephson voltage standard for example). From there, there is the concept of traceability and NIST certificates and so on, so that you can transfer the standard from one device to another.
If you want some search terms, google "electrical metrology" which is the field encompassing this, and "Josephson junction" which is the modern standard of a volt. These devices are maintained by NIST in the US and other orgs worldwide, that maintain a traceable chain of standards that is used to calibrate devices.
@@supernumex May I add that the traceability does not guarantee the accuracy of any unit, it does only guarantee that if a unit used to test your unit, is found faulty, is it possible to warn you about the error.
Every test gear used from the highest level and to user level is within a margin of error that gets transferred to the unit tested by it. One more thing is that most units do not get calibrated as in adjusted in accuracy but is tested for deviation and change in accuracy from new to now. Often is it more important to know how your unit changes over time than try to bring it back.
It comes from very accurately instruments (transfer standards), which in turn are calibrated using some sort of primary standard at a standards lab, for instance the national metrology institute of the respective country. For voltage, that primary standard could e.g. be based on a Josephson junction: en.wikipedia.org/wiki/Josephson_voltage_standard
@@friedmule5403 Very good points.
Mooi Apparaat
Where is the Teardown??😰
Hi Shahriar. Such a great review. Would you be able to compare this to a much cheaper instrument like a DER EE DE-5000 for us would cannot immediately afford a the better E4980A.
I have been meaning to get one.
👍👍👍💖👍👍👍
Either people really want to hold on to their 4339Bs or they didn't make a lot of them.
Then if you find one you see they REALLY want to hold onto them when you see the price! You might have better luck finding a 4275, which is a 1980's LCR meter that goes to 10 MHZ. It is a LOT larger physically (the 19" rack box that is typical of HP instruments of that era) than these new instruments but it has comparable performance. I got mine in working condition off ebay for $600.
You may want to consider an Advantest R8340A as an alternative. It combines a HV DC source and a low current meter to make a very high resistance meter. They are much cheaper on ebay and still have good performance. I have one and have had good luck with it.
Fantastic instrument if you have close to 20 grand to buy one! I am amazed at how you manage to get literally millions of dollars worth of instruments! Super rich patrons? You won the lotto? Sold your soul? :)
goot
Do a giveaway and send me this baby I need for really important science stuff
اقا دمت گرم یه سری اموزشی برای طراحی رادیو هم بزارید
Hello World
Second
Werd mal erwachsen ;-)