Simple fix to get rid of most of the switching noise. It's a possible scenario that the power supply was originally designed to use different mosfets with much higher gate capacitance, but then the manufacturer bought a cheap truck load of the Fairchild mosfets and used those instead.
What is a good name brand mosfet and better gate capacitance, so it may be better to swap it instead of modding. I just ordered a 'qw-ms3010d 'dc power supply and i want to start learning electronics for my flashlight hobby.
@@forrestgumpv9049 It's not so much about the brand name, but there are a lot of mosfets out there and it's a jungle, so trying to find one that would work is quite a task. Anyway make sure the voltage and current rating is high enough and then look at the Qg parameter usually in nano Coulomb and some datasheets will show Ciss as the input capacitance in pico Farad. Cheers, Jake
Completely different Power Supply, same problem, same solution. Thanks for pushing me into the right direction. I realy like the way of how you put your videos together. No mocking about, no endless talking about the same thing hundrets of times over... Straight to the point, boom, live with it. Like it.
The probe "antenna" along with your surmising the problem was overdriven mosfets and your solution represent an unusually high skill level. Most excellent! Thank you.
Interesting, well it's good when a manufacturer is prepared to listen and make changes to improve their products. Thanks for letting me know. Cheers, Jake
I've never seen that trick with the loop on the probe before. I'm chasing down some HF noise in a completely different context and that will be really useful, thanks!
With very little load on the power it was jamming my icom 7300 ham transceiver and putting large pulses on some TTL logic signals I was testing. Your corrections fixed the problem. Thank you!!!
Great work, the videos don't get any better than this. You are a very good teacher. I will do this mod to my pwr supply. Thank you. I couldn't read the part number of the cap, you might consider posting a parts list in future. Thanks again.
If you experience the same kind of noise on another switching power supply, then there is a good chance it will work, but of course I can't guarantee anything. Cheers, Jake
That's a good tip (ha) on using the ground wire of the scope probe to generate a pickup for pinpointing noise. I always used a few loops of wire but did not think of using the probe wire itself. Nice job rooting out the noise source. Adding the RC filtering helped significantly making the supply a bit higher in output quality. Cheers,
I was considering buying an LW-K3010R then I saw your earlier video showing how noisy the output is so I changed my mind. This video made me change my mind again and I've just ordered one! Thanks for both videos. I've subscribed, liked, and I'll be watching your other videos after this as I like your style and content. Good luck with this channel.
Thanks, I'm happy to hear my review and update fix was useful. It's of course not the worlds best power supply but I do think it's good value for money, especially with the noise fix implemented.
Thanks, it is a real pleasure to see such intelligent step by step analysis and solution development. I also liked your EMI detection tip. Yes, my degree is from 1972 so I too must learn SMD soldering skills, hope my eyes will be up to the task!
Thanks, much appreciated. When I switched to SMD I thought it would be much more difficult, but actually I generally find it easier and quicker to solder SMD components today compared to through hole. Cheer, Jake
@@TheStuffMade - On your pickup coil for the scope... I use one of those old suction cup coils used to record phone calls from the receiver or handset of a telephone. Those are 500 or so ohms and are very sensitive with good output to find BS in circuits. Also they have great bandwidth for audio signals. I have used them to detect whether coils are doing their job correctly. One of the more interesting uses was for vector monitor yokes. There is a substantial difference in the fields between a good one and a bad one. Some electronics makes really crazy noises if you have a listen with that coil also.
Just ran across this and I am happy to find a solution. By any chance, did you use the FFT function to discover the frequency of those spikes? The RC values show a cutoff Freq. of ~102Hz. Just curious. A full bridge produces 120Hz ripple so is that consistent with the switching IC? Thanks for taking the time to make the video!
Great video and something that LongWei should think about doing to improve their product. I can't imagine it would affect the cost all that much and it would bring the rating of their product up over others within the same budget market. Wish I could tag them in this comment... :)
Thanks, I believe the engineers who designed the power supply probably had designed it correctly, then the manufacturer changed some of the parts to save a few cents. But yeah, I wish they would implement the tiny mod to get rid of the awful noise. But they've probably already made 1000s of these stocked in warehouses so it's unlikely they'd consider making any changes.
Wonderful video. As a MSc Electrical Engineering from 1998 I got to 'update' my SMD soldering, but this is gonna work. For reference for others: the components and their respective values you ended up with are at 14:47
Apologies if this is a dumb question, but for the switching noise and ripple to be present on the output, that would mean that it’s severity exceeded the level of smoothing provided by the specification of the main output capacitors? Put simply it’s so bad that the hard-switching peaks are making it through the output filtering?
It's not a dumb question, but switching noise and ripple are two very different types of noise and must be dealt with in different ways. You can lower the ripple by upgrading the output filter, increasing the switching frequency or adding more phases etc. With switching noise you have to eliminate it at the source as it's EMI (I believe I mentioned this in the video). In this PSU it's generated by the MOSFETs because of excessively fast switching. Hope it makes sense. Cheers, Jake
@@TheStuffMade Ah, I was thinking about it the wrong way still even though yes you did mention it being EMI. Which is why using the probe as a EM receiving antenna was effective, I see now. The conductors connected to the mosfets act as transmitting antenna for the harsh HF switching transients, which can be induced into the conductive circuitry of any devices or wiring close enough in proximity. (receiving antenna) Or at least this is way I understand it to be, as I’ve been learning the importance of reducing unwanted EMI, especially as of late, since I’ve been getting into the shortwave and CB hobbies. And the somewhat new and acute awareness that RF is generated by every device that has switching circuitry, which is not always dealt with or suppressed to a level that doesn’t cause unwanted RF interference in certain frequency ranges. Harmonics. Inductance. It’s a lot, but I’m been steadily learning, and slowly small levels of understanding come here and there as I poke around and try to familiarize myself.
Thank you for the mods. Did you notice that the power switch only turns the power off to the bridge rectifier but not associated mains input filtering circuit , also the metal case is not earthed.
Thank you for sharing your opinion of this power supply and especially the noise reducing mods. My PSU arrived today so I'll order the parts as I know I don't have the resistors and I'll have a look through my caps. Great channel, so subbed, thanks.
Ive recently bought an RS310p 30V 10A PSU. Ive bookmarked this video and will do the same mod to the MOSFETS in mine. I'll also add an inductor between the board and output terminals.
A big thanks for the excellent video. I've risked a Topshak (BG's own rebrand) LW-3010EC and hoped that the core is the same (from the LW- type) and yes it is so I did the same mod (actually with 2 of them after seeing the result of the first one and how reasonable the price was) with very similar results (although used 27 Ohm 0805 as that was in my drawer and didn't want to push it too hard as for really low noise I have linear PSU). RMS went down from 12.7 to 3.7mV and Pk-Pk from 281.1 to 83mV in a noisy environment (checked with DSO, tried to avoid noise leaking in as much as possible). One issue though: I guess it's not just about more heat but also reaction time for quick load change that has increased as well. None of my concern, just perhaps something that's also good to be aware of.
Thanks, good to hear you got rid of the noise. These switch mode power supplies are very slow reacting to changes in load. See this video I did before the mod: th-cam.com/video/RdT-xWC4h4k/w-d-xo.html The mod itself will not have any effect on the power supply's reaction time, it just takes the edge of the MOSFET switching. Cheers, Jake
OK, it's even better then (and I was wrong about reaction time but I didn't look into it much as I have linear PSUs as well that I use for more critical tasks; it's also somewhat similar to what you also have but in a cheaper/simpler edition). Thanks again for the video, it has helped me a lot to make this fix quick and easy and also for the other videos that explore PSUs in a way that I rarely see but found very useful.
Hi nice video ! but for me it looks risky to tune it without analyzing the pulses on the gates, to see the deadtime and also to evaluate the increase of heat of these transistors ! i'm curious to see the curves of gate signals and how much this rc filter is slowing down it. how about a correct current loop for the transformer, transistor and capacitor with also a shield ?
Hi, thanks for the great video! I got a new Longwei PDS-3010G. I think it's probably quite similar to the LW-K3010D and I'd like to improve the noise/ripple. One difference I see is that the MOSFETs B22 FQPF 10N60C look more like "fakechild". There is no Fairchild logo on them. That should work anyway, right? I'm not very experienced. Thanks a lot!
I just opened my LongWei PS-3010DF to compare it with the K3010D in this video and it's almost identical. When putting everything back together, I was testing the ground connections (front ground banana socket and back AC line connector socket) to verify continuity and noticed that the metal chassis is not grounded. A bit concerning, considering that the incoming live wire is only soldered to the underside of the PCB (the PCB lays flat in this model, not vertical) and close to the chassis bottom.
It's a good thing to check 👍a shame they don't pay more attention to basic safety. I'd prefer the ground fault relay doing it's thing instead of getting shocked.
So looking at the new version of this powersupply, looks like longwei fixed the noise issue using a very similar method. I was going to do the mod but saw that there was a 10 ohm resistor on the gate of the MOSFETs. No extra capacitors though.
@@TheCasualDrinker Noise is still present in the supply, and it may not be suited to work with noise sensitive devices, however, it is not too bad. The noise is definitely better than the old versions though.
@electrifyingvids3545 i'm a bit of a HAM man so noise isn't the greatest thing to have. however I wonder if adding capacitors and the resistors aswell might help even more ?
Thanks , fantastic video for people who do not have the practical experience and are not familiar with SMPS ! I have learned the trick of "coiling" the scope's ground wire (making it an Antenna). The manufaturer of this SMPS should take a look at this video :-)
Thanks, always happy to hear you enjoyed the video. Agree, I wish the manufacturer would put a little more effort into their product, it was probably designed to perform better but then they changed some components to save a few cents. Cheers, Jake
Very practical and educational. I just wonder why the manucafturers do not bother to put this output filtering right away? that would probably cost like 50 cents more per PSU really. And they could brag that their switching PSU produce low level of noise to attract more buyers.
Thanks, it is a likely scenario that the power supply was designed to use other components, but the manufacturer changed some parts (e.g. the MOSFETs) to save money but at the same time degraded the performance. These lesser known brands usually have their marketing department make up the specs anyway and they've probably already got warehouses full of them so they only really care about getting them sold asap. Cheers, Jake
Thanks, I hope it works out, I have had reports that this mod also worked on other power supplies, but I can of course not guarantee anything. Cheers, Jake
what about the improvement of the snobbers instead of slowing down the gate, we could perhaps avoid the loss of thermal efficiency (noob's question), good video thx
Thanks, well there is no snubber filter, so that's that. From what I see there is only some clamping to prevent exceeding the max Vgs. Anyway, not a bad question, if you own one of these, feel free to experiment, just don't electrocute yourself. :)..
Hi ! thank you so much for this excellent video I subscribed immediately and i will watch all other video on power supplies, my main interest these days. I am a beginner trying to learn something. I have some questions. 1) in order to understand better the cheracteristics of noise would not be better to check with a spectrum analyzer ? to see the noise vs. frequency ? 2) as i am looking for very low noise voltage sources for some audio preamps i am putting together ... which would it be your first choice for a very low noise voltage source ? linear or switching ? and in case of a linear PSU do you think that IC regulators like the common LM317 can get down ripple to some mV or a more complex design is needed ? In my testing i have been amazed by the difference i can listen from a same preamp using different power supplies. Incredible. With higher noise PSU the sound is confused muffled ... with lower noise one is crystal clear and very nice. I am using a LM317 regulator at the moment bought on ebay. But i am very willing to try something else even less noisy. Thank you very much again. Kind regards, gino
Top draw modification. Great presentation. I was looking to get the LongWei 3010E with the extra pre-set and lock button functionality. Looks great until I read the Ripple and Noise specs; CV≤30mV (effective value) ; CC≤20mA (effective value) - I guess 30mV rms ! and 20mA rms ! Really ? Hoping the circuit is similar to your modified LW-K3010D, as apart from the dreadful noise issue the rest of it looks good value. Do you have any familiarity with the higher spec 3010E please? Keep up the good work.
Thanks, I don't have any experience with the LongWei 3010E, it looks nice and as you say it's got all the features missing from the 3010D. It's hard to say if the switching circuitry will be similar, the LongWei name might just be a sales branding and the power supplies could've been designed and manufactured by different companies. Since the cost is around the same as the Korad ka3005d, you should have a look at that power supply, it's only 5A max, but it's a low noise linear design, I've had a Korad for quite a few years and I'm very happy with it's performance. Cheers, Jake
@@TheStuffMade Jake. Thanks for your considered response. I took a look at your other comparative test video ' Homemade vs Korad vs Lonwei' and I take your point that the Korad is a cleaner and safer bet than the LongWei.The LongWei seems to have a nasty sting on some of the tests. I am developing Arduino based circuitry, so I appreciate you steering me towards the Korad. The 30/10 version is expensive, but maybe the 30/05 Korad would be fine for most applications - hopefully even the heavy load 12V Peltier systems. Many thanks. John
I have the same exact scope as yours. I wonder if there is a firmware update available for it? Nice video, I liked the probe loop technique to detect noise. Thanks.
Thanks, I updated the firmware on mine years ago, it greatly increased the update speed. They might still have updates available for download on their website.
Thank you so much for everything. I live in Viet Nam, longwei power supply is the best deal for me now. How does the bench power supply work in this year after noise mod?
Just have bought one of them, thank you it will be very useful for me in the future! Do you think this could be a problem for working with audio frequency signals? At which frequency does the noise really become a problem?
The noise is mostly a problem when you look at small signals on a scope as they will drown out in the switching noise. For audio work I would recommend using a linear power supply except perhaps if you work on projects like class D amplifiers.
@@TheStuffMade unfortunately linear supply are big and less cheaper than SMPS.. maybe in the future I would build one by myself, as it also seems not very difficult, thank you very much!
@@lelesecchi6140 Sure I understand, you can still use the Longwei just fine. I recently did a video on building a small and simple dual rail linear power supply ideal for low power audio projects if you are interested.
Hi sir, I need your help, In doing the mods I did a mistake in the bottom mosfet for an error indeed of a 22 ohm resistor I put a 22k. During the test at 12V everything seems working but when I push the voltage to 30V I got short and the house protection cut down, after verifying my fault I dismounted the Mosfet and verified that its pins are alla shorted out, I fear can be more component to check, the 22k resistor on the source is ok what else should I have to check other than the mosfet that is broken for sure? Thanks
Hi, first check the fuse in the power supply, I'm surprised if that didn't blow before your mains fuse kicked in. Besides that a lot of things could go wrong if one of the mosfets shorted out, check the transformers for continuity. I'd expect you'd be able to smell or see some damaged components inside. It will probably not be worth trying to repair it if any of the transformers are faulty. Cheers, Jake
@@TheStuffMade Ho thanks for your answer, it's the first thing I've checked out the fuse it'ok, I cannot did more testing since as soon as I power it on the home protection (magnetotermic) kick in but fuse will not blow out this is really strange. I ordered a batch of Mosfet to test out. Something in your mind that should I order for a component swapping activity?
@@marcovalentini1845 It's probably a ground fault causing your mains relay to switch off, something is likely shorted to mains ground somewhere. I doubt it's worth the effort to try repair it especially not if you don't have experience n repairing switch mode power supplies, but I'd check all the transformers for shorts and continuity first. Then usually a visual inspection and a smell test of all the components will give you some idea what might have failed. Cheers, Jake
@@marcovalentini1845 Understand and it's a good way to learn by repairing things. Just be careful and don't connect it to the mains until you are confident it's been repaired. The safe way to diagnose and repair a switch mode power supply is to inject low voltage from another power supply to power up the circuit. That way you can power up the secondary low voltage circuity while limiting the current so no additional damage it done and start doing measurements to find faults. But you'll need tools like a scope to see if you can get the switching going. Cheers, Jake
hello there, Thank you very much. I have 2 of that same PS only one does this the other one is very quiet. BG (big prase here) is sending a replacement, it was so noisy I could not test anything with my scope because of the noise this PS putout the trigger would just not lock. if the replacement does the same thing I will do this. Greate fix.
Nice video, I was wondering if a 1/4w or 1/2w through hole resistor instead of a 0603 would achieve the same effect, or if they would be less effective somehow? I don't have any SMD 33Ohm resistors on hand currently. Thanks
Thanks, yes you can use 1/8 or 1/4 watt through hole resistors instead of 0603, you'll just have to scrape off a bit more of the solder mask to solder them onto the traces safely. Otherwise you can buy 0603 very cheap online something like $1 including shipping for 100 pieces. Cheers, Jake
I have 22R and 47R 0805 250mW resistors but no 33R. Based on your previous testing, do you think it would be better to go with the 22 or the 47? Sorry, I know this was 3 years ago already and you may not recall all the details!
does this mod work the same at high power? I also noticed in your old review video on this bench PS that it ha difficulty in regulating in the 20V range. did this mod solve that problem too?
Thank you very much for the explanations, impressed by your knowledge. I know that I am a bit abusive in putting it into this topic, since the topic that I am going to indicate to you differs from the one presented. I describe a problem I have with a system based on an ESP32 microcontroller. After many programming tests that I carried out in the house where I live, I got a stable system without resets that works well, but when I moved it to its destination to my country house to control irrigation, boiler, hot water, heating, mains water pressure, temperature... etc. , I discover that the system resets randomly many times a day. The only explanation I can find is that the mains current reaches my house with an unusual noise. I have tried different power supplies for the ESP32 that works at 3.3V and the differences are minimal, and it is independent if I keep the ESP32 disconnected from all the sensors and relays, the problem is always the same. Could you tell me what I could do to eliminate the noise that surely makes my system not work continuously, thank you very much for your help and patience.
Hello, I don't think I'll be able to help you in the youtube comments, but I recommend you try post on the eevblog forum, they're very helpful and it's likely someone will solve your problem. Cheers, Jake
Nice video. I am curious when adding the caps to the Gate, I would think if the Gate is not 'discharged fast enough then you can have MOSFET's on at the same time in this half-bridge setup unless the dead time is long enough and not causing the issue. I also do not see they put on any ferrite beads on the MOSFET's legs to reduce switching noise.
Thanks, if you look at the design you'll see the gate signals go though a transformer from the controller on the secondary side, so a ferrite bead wouldn't make any difference. Cheers, Jake
@@TheStuffMade That is typical Gate drive transformer topology to isolate cold side from hot side of the circuit, the Ferrite is on the MOSFET legs is typical to reduce ringing and radiate switching noise so it will pass compliance test for inject and radiate noise, we had to do lots of stuff, I.E. board layout, wiring, shielding, etc. in our power supplies we design, the unit you have probably does not any compliance certificate.
@@budmartin3344 My point was more that if a ferrite bead is primarily going to take the edge off then the transformer would've already done that. But yeah, I agree this power supply would never pass any compliance testing. It's likely it was designed to perform better but then components were replaced to keep production cost at a minimum. Cheers, Jake
Why adding an extra capacitor instead of increasing the gate resistor? It's R x C that counts (where C = Cgate + Cextra), so you could just increase R and leave the extra capacitor, no?
Great video and very nice technique in breaking the tracks to add the resistors. It might have been interesting to see the same test with ferrite beads rather than resistors.
Thanks, these are bipolar film capacitors, the polarity doesn't matter. As for shielding I assume you mean where to connect the side of the capacitor that's got the outside film, then in this case it doesn't matter because both sides of the capacitor are connected to low impedance. This only matters in very high impedance circuits with small signals where you're trying to keep as much noise out as possible, like tube radios etc. Cheers, Jake
30 ohm resistors and 47nF caps. 2 of each. I bought a larger power supply (link in reply below), 120VDC and 3amp, 360W. The 47nF caps shown are only rated for 100V, is this mod going to work for me I think I need higher voltage caps? Can I directly replace the ones in the video with 630V Polypropylene Safety Plastic Film 630V 47nF?
Hello!! About 1,5 years ago, I bought through Banggood this product LW-K3010D. Since last week, I started having a problem each time I try to raise the voltage from 0-30V on the POTENTIOMETER. It does not exceed 9.2V on the screen, but also at the output of the connectors. Since it does not work properly, I cannot use if for my job. Can you please advise me on how I could fix it? Waiting for your prompt reply. Thank you!
Hi Sakis, this is not something that can be repaired in a few comments. Your best bet is to take it to a local shop and see if they can repair it. Cheers, Jake
Thanks Jake for your reply!! Unfortunately here in my city there are not many workshops that can help me! I have also sent the Company (LONGWEI) 2 emails but.....they haven't sent me anything!! I don't know what else I can do. Too bad it can't be repaired!!
@@Markoulis66 Hi Sakis, yeah, it's a shame. A repair shop would probably need to charge as much as the cost of a new one, it can easily take a few hours to diagnose order parts, replace parts, then test to make sure it works again. So unless you can do it yourself or know a friend, then it's probably cheapest just to buy a new one. Cheers, Jake
I have the same model. It just died on me. Zero amp on the display and it doesn't charge anything. I've opened it up and I don't see any components that are blown. Do you know what it might be the problem?
Good Day sir, i have same model but the issue is when i increase the output it doesn't increase more than 8.9 volts. Do you have any idea on what could possibly the fault? I already reaplaced the multi turn pot. But the problem stays.
hey brother i need your help, i have this power supply but when i turn it on and turn the current knob the green light does not come on so i can't adjust anything, i can't figure out how to get it working
Brilliant work. My PSU died when I was using it to drive an arcing spark generator (don't ask), but I found the circuit diagram for the same PSU but with a less max ampage. So the area that you worked on failed. One of the large MOSFET's went short, 2 of the surface mount transistors that you pointed to went short, and so did ones of the schotky diodes next to those surface mount transistors - I've replaced all those . The PSU currently pulses the display showing 000 when switched on. Now I'm wondering if the switching controller you pointed to has died? I think this is the circuit diagram I had (thought you might find it interesting or useful) = www.eevblog.com/forum/repair/permanent-cc-and-no-output-(longwei-k3010d)/
the same thing to me with the short, but I also changed the U5 TNY820 AND THE PC817B PHOTO COUPLER, IT HAS RESUMED WORKING, PULSA X WHICH IS PROTECTED THROUGH U4 SG3525.
what is the reason for removing the noise? i am thinking of powering a radio transceiver that requires "12v", i have not done this yet, but would this noise come through the radio some how?
Hi, the switching noise if high frequency and have a broad bandwidth, it will radiate out into everything. If you try to measure something on a scope it gets very annoying because you don't know if the noise is from the device you're working on or it's from the power supply and it will make it impossible to make any kind of sensitive measurements. It will also create radio interference, as in all kinds of radio equipment will be affected, even if they're not powered from the supply, it's enough the supply is just switched on. Generally switching noise is very undesirable, especially in any kind of electronics lab environment. Cheers, Jake
Noise is bad for sensitive circuits, coms, medical, etc. Switching power supplies are inherently noisy. Some are really noisy. Linear PSU's are much cleaner. Generally, less noise is better.
@@sekboi I have that same PS. BG is sending a replacement, it was so noisy I could not test anything with my scope because of the noise this PS putout the trigger would just not lock.
Hi sir, i have the same power supply and the problem is that cannot keep Amps stable, up - down. Could you tell me sir where to look or what to replace?
Hello, the output current depends on the current limit and what load you have on the power supply. To set the current limit you need to set the voltage to something e.g. 5V then short the output and you can adjust the current to what limit you need. Hope that helps, Cheers, Jake
@@TheStuffMade i understand what you said sir it's my mistake, i didn't said it right. What i m just saying is, after set the voltage that i need, and adjust the current as you said on the display, initially is e.g. 2 Amps after a second it is 2,5 Amps then 3, 2, 2,5. it is increasing and decreasing by it self with stable voltage. What could i do?
In my opinion, the worst feature of this PS is the current spike when used as constant current supply at low levels. The behaviour reminds that of a line output with a relevant capacitor in parallel. Did you check if your mod do not worsen the already bad spike?
I have another video testing the speed and performance of this power supply: th-cam.com/video/RdT-xWC4h4k/w-d-xo.html It's typical for switch mode power supplies to react slowly to changes because of the way they work, most linear power supplies have superior performance in that regard, but switch mode supplies are strong when it comes to high power and lower cost. The noise fix will not improve the speed of the power supply.
Yes it will limit the noises because of decreasing dv/dt That means the mosfets switch slower(fall time and rise time increased) So the mosfets will be get hot and ...dead
Neither are particular dangerous to electronics as long as they are within reason, but you always want a bench power supply to be a low noise as possible otherwise it can affect the behavior of your circuit and make it very difficult to do any reliable measurements. Cheers, Jake
i have a minleaf 30v 10a power supply i havent got a scope to check it though well i have but its one of those mini DSO138 i dont think its capable of these tests but i could be wrong.
These little scopes are good up to 100-200kHz. They are useful for audio etc. but for these noise measurements, as it's very high frequency switching noise, you need a scope capable of 100-200Mhz, so about 1000x faster to get good measurements. Cheers, Jake
Hello David, you need the resistors, I didn't just add them for fun. If you don't add the resistors and only the caps you'll put way too much load on the driving circuitry and it'll likely let some magic smoke out. Cheers, Jake
I have a source of this LongWei LW-K3010D it is falling unregulated voltage does not let it go over 9V then turn on the sera fan q can be the mosfet???
Had the same issue. Probably a later revision board which now has a TL413 (U7) as a ref voltage. Check pin 1 as I bet it no longer maintains a constant voltage.
@@TheStuffMade about the transistor, I ask just for curiosity, and my power supply is the same as the video, but is not changing voltage, just still at 0, I try replace PWM (SG3525A), and CI 084C, and LM2904, I replace the transistors, check de potentiometer BUT NOTHING!! I’m really upset, some idea? thanks for attention
Hi how does the Korad ka3005d vs LW-K3010D with mod compare in noice level ? i would like to compare Vpp and Vmax. Since i can get the K3010D for 56 dolars and the Korad for 85 dolars, both with shipping. Since for my understanding the Fix, does reduce alot the Vpp.
The Korad is the better power supply even after the fix. If I could have only one of these power supplies, I'd get the Korad. However the Longwei is great to have when you need a lot of power as it's capable of delivering more than 300W output. Make sure shipping is included when you compare the cost as the Korad is a linear power supply with a large transformer, so it's heavy and often cost a lot to ship. Hope this helps, Cheers, Jake
Hi Marco, it's 47nF not uF, use a film capacitor. The capacitor doesn't see any high voltage so e.g. a 50V rated capacitor is fine, or use one of the green ones I use, they are very common and I believe they are rated at around 100V. Cheers, Jake
![BOWKYLION - วิงวอน (ex-change) [Official MV]](http://i.ytimg.com/vi/cLdnZWDaO-w/mqdefault.jpg)
Simple fix to get rid of most of the switching noise. It's a possible scenario that the power supply was originally designed to use different mosfets with much higher gate capacitance, but then the manufacturer bought a cheap truck load of the Fairchild mosfets and used those instead.
What is a good name brand mosfet and better gate capacitance, so it may be better to swap it instead of modding.
I just ordered a 'qw-ms3010d 'dc power supply and i want to start learning electronics for my flashlight hobby.
@@forrestgumpv9049 It's not so much about the brand name, but there are a lot of mosfets out there and it's a jungle, so trying to find one that would work is quite a task. Anyway make sure the voltage and current rating is high enough and then look at the Qg parameter usually in nano Coulomb and some datasheets will show Ciss as the input capacitance in pico Farad.
Cheers,
Jake
Thanks I finished the mods and ripple reduced to few mv
Completely different Power Supply, same problem, same solution. Thanks for pushing me into the right direction.
I realy like the way of how you put your videos together. No mocking about, no endless talking about the same thing hundrets of times over... Straight to the point, boom, live with it. Like it.
Thanks!
@@TheStuffMade Excellent, background noise is a big problem in my bench power supplies.
Next week I try to mod one at least.
Best Regards.
You went a Longwei to making the power supply almost perfect!
lol... very good, thanks.
The probe "antenna" along with your surmising the problem was overdriven mosfets and your solution represent an unusually high skill level. Most excellent! Thank you.
Thanks!
I just recently bought one of these units and to my surprise discovered that your fixes where already part of the newer revision PCB!
Interesting, well it's good when a manufacturer is prepared to listen and make changes to improve their products.
Thanks for letting me know.
Cheers,
Jake
I've never seen that trick with the loop on the probe before. I'm chasing down some HF noise in a completely different context and that will be really useful, thanks!
I love when you busted out that book of resistors. That was boss!
Available at Amazon 👍
With very little load on the power it was jamming my icom 7300 ham transceiver and putting large pulses on some TTL logic signals I was testing. Your corrections fixed the problem. Thank you!!!
This is so educational and wonderful use of the scope to track down the issues. Thanks
Nice tip on how to use scope probe to pinpoint noise source. Video was way worth viewing just for that. THANX...
Great work, the videos don't get any better than this. You are a very good teacher. I will do this mod to my pwr supply. Thank you.
I couldn't read the part number of the cap, you might consider posting a parts list in future. Thanks again.
Thanks, they are 100V rated film caps, you can get a pack of them for $1 on ebay, aliexpress, amazon etc.
Cheers,
Jake
Just implemented this modification. It works absolutely great. I was really getting annoyed with all the noise of the PS but this solved it. Thanks!
Hi Paulo, that's great feedback, good to hear it worked for your PSU as well.
Thanks - Jake
Done. 400mV noise before, 10mV after the mod. Great.
Lovely!! I wonder if we can replicate this process on other brands of cheap bench power supplies
If you experience the same kind of noise on another switching power supply, then there is a good chance it will work, but of course I can't guarantee anything.
Cheers,
Jake
Cheers! Bought this and did the mod straight outta the box (albeit with 0805 resistors!)
LPF with a corner frequency of 102.6KHz.
Nice, always good to hear another Longwei has been denoised. I can report that mine still works just fine.
Cheers,
Jake
I love the quick and dirt EMC probe. Love it ! Great video !
That's a good tip (ha) on using the ground wire of the scope probe to generate a pickup for pinpointing noise. I always used a few loops of wire but did not think of using the probe wire itself.
Nice job rooting out the noise source. Adding the RC filtering helped significantly making the supply a bit higher in output quality.
Cheers,
Thanks, yeah it turned out better than I'd hoped for.
Really helpful videos, thank you for these. I was casting around for a cheap 10A power supply and this convinced me to go for this one.
I was considering buying an LW-K3010R then I saw your earlier video showing how noisy the output is so I changed my mind.
This video made me change my mind again and I've just ordered one!
Thanks for both videos.
I've subscribed, liked, and I'll be watching your other videos after this as I like your style and content.
Good luck with this channel.
Thanks, I'm happy to hear my review and update fix was useful. It's of course not the worlds best power supply but I do think it's good value for money, especially with the noise fix implemented.
Very good investigation of the problem, and explanation of your fix. I have one on order from BG, so will see if I need to do it when it arrives.
Thanks Robert, the power supply will work fine without the fix, but if you need it to be less noisy it's an easy option.
Thank you, good sir!
I'd never have found the source of the issue on my own. Now I can get down to fixing it.
Thanks, I can report the mod still works great, no issues.
I ordered the exact same device. Will try this mod when it arrives. Thank you for the wonderful vide.
Thanks, it is a real pleasure to see such intelligent step by step analysis and solution development. I also liked your EMI detection tip. Yes, my degree is from 1972 so I too must learn SMD soldering skills, hope my eyes will be up to the task!
Thanks, much appreciated. When I switched to SMD I thought it would be much more difficult, but actually I generally find it easier and quicker to solder SMD components today compared to through hole.
Cheer,
Jake
@@TheStuffMade Thanks Jake,
You motivated me to order some SMD resistors 😊
Best wishes
Mark
@@TheStuffMade - On your pickup coil for the scope... I use one of those old suction cup coils used to record phone calls from the receiver or handset of a telephone. Those are 500 or so ohms and are very sensitive with good output to find BS in circuits. Also they have great bandwidth for audio signals. I have used them to detect whether coils are doing their job correctly. One of the more interesting uses was for vector monitor yokes. There is a substantial difference in the fields between a good one and a bad one. Some electronics makes really crazy noises if you have a listen with that coil also.
Wow. Great. It shows the in depth knowledge on how SMPS. Thank you.
Thanks very much for sharing this mod, have just done this to my unit and its made a massive improvement too!
Thanks, happy to hear you had success with the mod.
Nice mod! And also clever home made probe to scan for EMI 8-)
Great Video ... just got hold of the same power supply ... may attempt this fix one day myself ... thanks again ...
Thanks
Just ran across this and I am happy to find a solution. By any chance, did you use the FFT function to discover the frequency of those spikes? The RC values show a cutoff Freq. of ~102Hz. Just curious. A full bridge produces 120Hz ripple so is that consistent with the switching IC? Thanks for taking the time to make the video!
Very well done, for switching power supply I think it is the best solution.
Thanks, I can report it still works well, no issues with this mod.
This was very helpful. Thanks for the great video.
Great video and something that LongWei should think about doing to improve their product. I can't imagine it would affect the cost all that much and it would bring the rating of their product up over others within the same budget market. Wish I could tag them in this comment... :)
Thanks, I believe the engineers who designed the power supply probably had designed it correctly, then the manufacturer changed some of the parts to save a few cents. But yeah, I wish they would implement the tiny mod to get rid of the awful noise. But they've probably already made 1000s of these stocked in warehouses so it's unlikely they'd consider making any changes.
Wonderful video. As a MSc Electrical Engineering from 1998 I got to 'update' my SMD soldering, but this is gonna work. For reference for others: the components and their respective values you ended up with are at 14:47
Thanks, much appreciated, happy you found the video useful.
Thank you for knowledge about switching power supply, you are great teacher.
Apologies if this is a dumb question, but for the switching noise and ripple to be present on the output, that would mean that it’s severity exceeded the level of smoothing provided by the specification of the main output capacitors? Put simply it’s so bad that the hard-switching peaks are making it through the output filtering?
It's not a dumb question, but switching noise and ripple are two very different types of noise and must be dealt with in different ways. You can lower the ripple by upgrading the output filter, increasing the switching frequency or adding more phases etc. With switching noise you have to eliminate it at the source as it's EMI (I believe I mentioned this in the video). In this PSU it's generated by the MOSFETs because of excessively fast switching. Hope it makes sense.
Cheers,
Jake
@@TheStuffMade Ah, I was thinking about it the wrong way still even though yes you did mention it being EMI. Which is why using the probe as a EM receiving antenna was effective, I see now. The conductors connected to the mosfets act as transmitting antenna for the harsh HF switching transients, which can be induced into the conductive circuitry of any devices or wiring close enough in proximity. (receiving antenna)
Or at least this is way I understand it to be, as I’ve been learning the importance of reducing unwanted EMI, especially as of late, since I’ve been getting into the shortwave and CB hobbies. And the somewhat new and acute awareness that RF is generated by every device that has switching circuitry, which is not always dealt with or suppressed to a level that doesn’t cause unwanted RF interference in certain frequency ranges. Harmonics. Inductance. It’s a lot, but I’m been steadily learning, and slowly small levels of understanding come here and there as I poke around and try to familiarize myself.
Thank you for the mods. Did you notice that the power switch only turns the power off to the bridge rectifier but not associated mains input filtering circuit , also the metal case is not earthed.
Thank you for sharing your opinion of this power supply and especially the noise reducing mods. My PSU arrived today so I'll order the parts as I know I don't have the resistors and I'll have a look through my caps. Great channel, so subbed, thanks.
Ive recently bought an RS310p 30V 10A PSU. Ive bookmarked this video and will do the same mod to the MOSFETS in mine. I'll also add an inductor between the board and output terminals.
I can't guarantee it will work on other power supplies, but good luck.
Cheers,
Jake
A big thanks for the excellent video. I've risked a Topshak (BG's own rebrand) LW-3010EC and hoped that the core is the same (from the LW- type) and yes it is so I did the same mod (actually with 2 of them after seeing the result of the first one and how reasonable the price was) with very similar results (although used 27 Ohm 0805 as that was in my drawer and didn't want to push it too hard as for really low noise I have linear PSU). RMS went down from 12.7 to 3.7mV and Pk-Pk from 281.1 to 83mV in a noisy environment (checked with DSO, tried to avoid noise leaking in as much as possible).
One issue though: I guess it's not just about more heat but also reaction time for quick load change that has increased as well. None of my concern, just perhaps something that's also good to be aware of.
Thanks, good to hear you got rid of the noise. These switch mode power supplies are very slow reacting to changes in load. See this video I did before the mod: th-cam.com/video/RdT-xWC4h4k/w-d-xo.html
The mod itself will not have any effect on the power supply's reaction time, it just takes the edge of the MOSFET switching.
Cheers,
Jake
OK, it's even better then (and I was wrong about reaction time but I didn't look into it much as I have linear PSUs as well that I use for more critical tasks; it's also somewhat similar to what you also have but in a cheaper/simpler edition).
Thanks again for the video, it has helped me a lot to make this fix quick and easy and also for the other videos that explore PSUs in a way that I rarely see but found very useful.
Nice to know before i buy one of these things what to do
Hi nice video ! but for me it looks risky to tune it without analyzing the pulses on the gates, to see the deadtime and also to evaluate the increase of heat of these transistors ! i'm curious to see the curves of gate signals and how much this rc filter is slowing down it. how about a correct current loop for the transformer, transistor and capacitor with also a shield ?
If it works it works and he did look at the mosfet temperatures…?
Hi, thanks for the great video!
I got a new Longwei PDS-3010G.
I think it's probably quite similar to the LW-K3010D and I'd like to improve the noise/ripple.
One difference I see is that the MOSFETs B22 FQPF 10N60C look more like "fakechild". There is no Fairchild logo on them.
That should work anyway, right? I'm not very experienced.
Thanks a lot!
I just opened my LongWei PS-3010DF to compare it with the K3010D in this video and it's almost identical. When putting everything back together, I was testing the ground connections (front ground banana socket and back AC line connector socket) to verify continuity and noticed that the metal chassis is not grounded. A bit concerning, considering that the incoming live wire is only soldered to the underside of the PCB (the PCB lays flat in this model, not vertical) and close to the chassis bottom.
It's a good thing to check 👍a shame they don't pay more attention to basic safety. I'd prefer the ground fault relay doing it's thing instead of getting shocked.
So looking at the new version of this powersupply, looks like longwei fixed the noise issue using a very similar method. I was going to do the mod but saw that there was a 10 ohm resistor on the gate of the MOSFETs. No extra capacitors though.
ive just oprdered a new one from amazon, Has your noise problem been solved from factory ?
@@TheCasualDrinker Noise is still present in the supply, and it may not be suited to work with noise sensitive devices, however, it is not too bad.
The noise is definitely better than the old versions though.
@electrifyingvids3545 i'm a bit of a HAM man so noise isn't the greatest thing to have. however I wonder if adding capacitors and the resistors aswell might help even more ?
It worked! Fantastic!
Excellent video. Looks like you used a mica capacitor. What voltage do you think would be appropriate for a 30 volt power supply like this?
Thanks, these a poly film capacitors, you can find them online at a few $ for a pack. I believe they're rated for up to 100V.
Cheers,
Jake
Wow amazing video. I hope I can apply same approach for my meanwell LRS-350
Thanks , fantastic video for people who do not have the practical experience and are not familiar with SMPS ! I have learned the trick of "coiling" the scope's ground wire (making it an Antenna). The manufaturer of this SMPS should take a look at this video :-)
Thanks, always happy to hear you enjoyed the video. Agree, I wish the manufacturer would put a little more effort into their product, it was probably designed to perform better but then they changed some components to save a few cents.
Cheers,
Jake
Very practical and educational. I just wonder why the manucafturers do not bother to put this output filtering right away? that would probably cost like 50 cents more per PSU really. And they could brag that their switching PSU produce low level of noise to attract more buyers.
Thanks, it is a likely scenario that the power supply was designed to use other components, but the manufacturer changed some parts (e.g. the MOSFETs) to save money but at the same time degraded the performance. These lesser known brands usually have their marketing department make up the specs anyway and they've probably already got warehouses full of them so they only really care about getting them sold asap.
Cheers,
Jake
Very impressed with this video presentation thank you very much.
Thanks!
i have a wanptek supply and the noise is insane !
I measured a ripple of over 4v @ 24v output no load.
A outputfilter could also help
Very useful video. I will attempt this on the Kungber 30v/10A supply. Am sure they would be somewhat identical.
Thanks, I hope it works out, I have had reports that this mod also worked on other power supplies, but I can of course not guarantee anything.
Cheers,
Jake
what about the improvement of the snobbers instead of slowing down the gate, we could perhaps avoid the loss of thermal efficiency (noob's question), good video thx
Thanks, well there is no snubber filter, so that's that. From what I see there is only some clamping to prevent exceeding the max Vgs. Anyway, not a bad question, if you own one of these, feel free to experiment, just don't electrocute yourself. :)..
Hi ! thank you so much for this excellent video I subscribed immediately and i will watch all other video on power supplies, my main interest these days. I am a beginner trying to learn something.
I have some questions.
1) in order to understand better the cheracteristics of noise would not be better to check with a spectrum analyzer ? to see the noise vs. frequency ?
2) as i am looking for very low noise voltage sources for some audio preamps i am putting together ... which would it be your first choice for a very low noise voltage source ? linear or switching ? and in case of a linear PSU do you think that IC regulators like the common LM317 can get down ripple to some mV or a more complex design is needed ?
In my testing i have been amazed by the difference i can listen from a same preamp using different power supplies. Incredible. With higher noise PSU the sound is confused muffled ... with lower noise one is crystal clear and very nice. I am using a LM317 regulator at the moment bought on ebay. But i am very willing to try something else even less noisy.
Thank you very much again. Kind regards, gino
Top draw modification. Great presentation. I was looking to get the LongWei 3010E with the extra pre-set and lock button functionality. Looks great until I read the Ripple and Noise specs; CV≤30mV (effective value) ; CC≤20mA (effective value) - I guess 30mV rms ! and 20mA rms ! Really ? Hoping the circuit is similar to your modified LW-K3010D, as apart from the dreadful noise issue the rest of it looks good value. Do you have any familiarity with the higher spec 3010E please? Keep up the good work.
Thanks, I don't have any experience with the LongWei 3010E, it looks nice and as you say it's got all the features missing from the 3010D. It's hard to say if the switching circuitry will be similar, the LongWei name might just be a sales branding and the power supplies could've been designed and manufactured by different companies. Since the cost is around the same as the Korad ka3005d, you should have a look at that power supply, it's only 5A max, but it's a low noise linear design, I've had a Korad for quite a few years and I'm very happy with it's performance.
Cheers,
Jake
@@TheStuffMade Jake. Thanks for your considered response. I took a look at your other comparative test video ' Homemade vs Korad vs Lonwei' and I take your point that the Korad is a cleaner and safer bet than the LongWei.The LongWei seems to have a nasty sting on some of the tests. I am developing Arduino based circuitry, so I appreciate you steering me towards the Korad. The 30/10 version is expensive, but maybe the 30/05 Korad would be fine for most applications - hopefully even the heavy load 12V Peltier systems. Many thanks. John
I have the same exact scope as yours. I wonder if there is a firmware update available for it?
Nice video, I liked the probe loop technique to detect noise. Thanks.
Thanks, I updated the firmware on mine years ago, it greatly increased the update speed. They might still have updates available for download on their website.
@@TheStuffMade I looked but couldn't find any. I will be thankful if you share a link or the firmware itself.
Thanks.
@@malgailany It was quite a few years ago, sorry I don't have a link.
@@TheStuffMade I just found one and updated it to firmware v3.8.
Link: files.owon.com.cn/software/upgrade/sds_upgrade_3.2_3.3_3.4.rar
Thanks a lot.
@@TheStuffMade Any chance you'll show us how to check the firmware version and update it? I can't find it so far, and the link below isn't working.
Thank you so much for everything. I live in Viet Nam, longwei power supply is the best deal for me now. How does the bench power supply work in this year after noise mod?
Hi, it's still working fine, besides the noise issue, that was fixed, I haven't experienced any issues with the power supply.
Cheers,
Jake
absolutely wild!
Just have bought one of them, thank you it will be very useful for me in the future! Do you think this could be a problem for working with audio frequency signals? At which frequency does the noise really become a problem?
The noise is mostly a problem when you look at small signals on a scope as they will drown out in the switching noise. For audio work I would recommend using a linear power supply except perhaps if you work on projects like class D amplifiers.
@@TheStuffMade unfortunately linear supply are big and less cheaper than SMPS.. maybe in the future I would build one by myself, as it also seems not very difficult, thank you very much!
@@lelesecchi6140 Sure I understand, you can still use the Longwei just fine. I recently did a video on building a small and simple dual rail linear power supply ideal for low power audio projects if you are interested.
Hi sir, I need your help, In doing the mods I did a mistake in the bottom mosfet for an error indeed of a 22 ohm resistor I put a 22k. During the test at 12V everything seems working but when I push the voltage to 30V I got short and the house protection cut down, after verifying my fault I dismounted the Mosfet and verified that its pins are alla shorted out, I fear can be more component to check, the 22k resistor on the source is ok what else should I have to check other than the mosfet that is broken for sure? Thanks
Hi, first check the fuse in the power supply, I'm surprised if that didn't blow before your mains fuse kicked in. Besides that a lot of things could go wrong if one of the mosfets shorted out, check the transformers for continuity. I'd expect you'd be able to smell or see some damaged components inside. It will probably not be worth trying to repair it if any of the transformers are faulty.
Cheers,
Jake
@@TheStuffMade Ho thanks for your answer, it's the first thing I've checked out the fuse it'ok, I cannot did more testing since as soon as I power it on the home protection (magnetotermic) kick in but fuse will not blow out this is really strange. I ordered a batch of Mosfet to test out. Something in your mind that should I order for a component swapping activity?
@@marcovalentini1845 It's probably a ground fault causing your mains relay to switch off, something is likely shorted to mains ground somewhere. I doubt it's worth the effort to try repair it especially not if you don't have experience n repairing switch mode power supplies, but I'd check all the transformers for shorts and continuity first. Then usually a visual inspection and a smell test of all the components will give you some idea what might have failed.
Cheers,
Jake
@@TheStuffMade many thanks, just worth trying for learning thanks
@@marcovalentini1845 Understand and it's a good way to learn by repairing things. Just be careful and don't connect it to the mains until you are confident it's been repaired. The safe way to diagnose and repair a switch mode power supply is to inject low voltage from another power supply to power up the circuit. That way you can power up the secondary low voltage circuity while limiting the current so no additional damage it done and start doing measurements to find faults. But you'll need tools like a scope to see if you can get the switching going.
Cheers,
Jake
hello there, Thank you very much. I have 2 of that same PS only one does this the other one is very quiet. BG (big prase here) is sending a replacement, it was so noisy I could not test anything with my scope because of the noise this PS putout the trigger would just not lock. if the replacement does the same thing I will do this. Greate fix.
Nice video, I was wondering if a 1/4w or 1/2w through hole resistor instead of a 0603 would achieve the same effect, or if they would be less effective somehow? I don't have any SMD 33Ohm resistors on hand currently. Thanks
Thanks, yes you can use 1/8 or 1/4 watt through hole resistors instead of 0603, you'll just have to scrape off a bit more of the solder mask to solder them onto the traces safely. Otherwise you can buy 0603 very cheap online something like $1 including shipping for 100 pieces.
Cheers,
Jake
@@TheStuffMade Thanks for the reply, I'll order some on Ebay and wait a few more days for a cleaner install.
Well done!
I have 22R and 47R 0805 250mW resistors but no 33R. Based on your previous testing, do you think it would be better to go with the 22 or the 47? Sorry, I know this was 3 years ago already and you may not recall all the details!
Hello, the 22R will probably work, but I'd recommend spend $1 to get the correct value.
Cheers,
Jake
@@TheStuffMade Ok, thanks for the advice. Will get the 33R 😁
does this mod work the same at high power? I also noticed in your old review video on this bench PS that it ha difficulty in regulating in the 20V range. did this mod solve that problem too?
Yes, and watch the full video.
Thank you very much for the explanations, impressed by your knowledge. I know that I am a bit abusive in putting it into this topic, since the topic that I am going to indicate to you differs from the one presented. I describe a problem I have with a system based on an ESP32 microcontroller. After many programming tests that I carried out in the house where I live, I got a stable system without resets that works well, but when I moved it to its destination to my country house to control irrigation, boiler, hot water, heating, mains water pressure, temperature... etc. , I discover that the system resets randomly many times a day. The only explanation I can find is that the mains current reaches my house with an unusual noise. I have tried different power supplies for the ESP32 that works at 3.3V and the differences are minimal, and it is independent if I keep the ESP32 disconnected from all the sensors and relays, the problem is always the same. Could you tell me what I could do to eliminate the noise that surely makes my system not work continuously, thank you very much for your help and patience.
Hello, I don't think I'll be able to help you in the youtube comments, but I recommend you try post on the eevblog forum, they're very helpful and it's likely someone will solve your problem.
Cheers,
Jake
Thanks I will ask there.
Thank you for you effort. Saved me a lot of time, my thanks
Thanks, always good to hear it was helpful.
Nice video. I am curious when adding the caps to the Gate, I would think if the Gate is not 'discharged fast enough then you can have MOSFET's on at the same time in this half-bridge setup unless the dead time is long enough and not causing the issue. I also do not see they put on any ferrite beads on the MOSFET's legs to reduce switching noise.
Thanks, if you look at the design you'll see the gate signals go though a transformer from the controller on the secondary side, so a ferrite bead wouldn't make any difference.
Cheers,
Jake
@@TheStuffMade That is typical Gate drive transformer topology to isolate cold side from hot side of the circuit, the Ferrite is on the MOSFET legs is typical to reduce ringing and radiate switching noise so it will pass compliance test for inject and radiate noise, we had to do lots of stuff, I.E. board layout, wiring, shielding, etc. in our power supplies we design, the unit you have probably does not any compliance certificate.
@@budmartin3344 My point was more that if a ferrite bead is primarily going to take the edge off then the transformer would've already done that. But yeah, I agree this power supply would never pass any compliance testing. It's likely it was designed to perform better but then components were replaced to keep production cost at a minimum.
Cheers,
Jake
Why adding an extra capacitor instead of increasing the gate resistor? It's R x C that counts (where C = Cgate + Cextra), so you could just increase R and leave the extra capacitor, no?
Hi Bigjim, it isn't quite the same when you look at the impedance of the circuit.
Cheers,
Jake
Thanks for the video, very informative, I would like to ask a question, do not understand the capacity of the capacitors used? thank you.
47nF 100V polyester film capacitors
@@em0_tion Thanks, I was looking for just that.
Great video and very nice technique in breaking the tracks to add the resistors. It might have been interesting to see the same test with ferrite beads rather than resistors.
Thanks, I don't think a ferrite bead would have much effect as the gate pulses are already going through a transformer from secondary to primary side.
Aren't the gate drive pulses amplified after the transformer?
@@userPrehistoricman There is no amplification, however it looks like they are doing some clamping, probably to keep it below the max allowed Vgs
Nice mod.
Thanks!
I just order one now. Was looking for a cheap model. Thx
would capacitor polarity (shielding) in this case matter? Thank you for putting these great videos out here
Thanks, these are bipolar film capacitors, the polarity doesn't matter. As for shielding I assume you mean where to connect the side of the capacitor that's got the outside film, then in this case it doesn't matter because both sides of the capacitor are connected to low impedance. This only matters in very high impedance circuits with small signals where you're trying to keep as much noise out as possible, like tube radios etc.
Cheers,
Jake
My power supply is : Jesverty sps-3010n, 30V 10A , with no load make a noise but when put load there’s is not noise how to fix it ?
Very helpful thanks.
best modification for 5 yaer of you tube EE univer's ty man
30 ohm resistors and 47nF caps. 2 of each. I bought a larger power supply (link in reply below), 120VDC and 3amp, 360W. The 47nF caps shown are only rated for 100V, is this mod going to work for me I think I need higher voltage caps? Can I directly replace the ones in the video with 630V Polypropylene Safety Plastic Film 630V 47nF?
I don't think the said voltage between gate and source exceeding 100v.
Hello!!
About 1,5 years ago, I bought through Banggood this product LW-K3010D.
Since last week, I started having a problem each time I try to raise the voltage from 0-30V on the POTENTIOMETER.
It does not exceed 9.2V on the screen, but also at the output of the connectors.
Since it does not work properly, I cannot use if for my job.
Can you please advise me on how I could fix it?
Waiting for your prompt reply.
Thank you!
Hi Sakis, this is not something that can be repaired in a few comments. Your best bet is to take it to a local shop and see if they can repair it.
Cheers,
Jake
Thanks Jake for your reply!! Unfortunately here in my city there are not many workshops that can help me! I have also sent the Company (LONGWEI) 2 emails but.....they haven't sent me anything!! I don't know what else I can do. Too bad it can't be repaired!!
@@Markoulis66 Hi Sakis, yeah, it's a shame. A repair shop would probably need to charge as much as the cost of a new one, it can easily take a few hours to diagnose order parts, replace parts, then test to make sure it works again. So unless you can do it yourself or know a friend, then it's probably cheapest just to buy a new one.
Cheers,
Jake
I had a similar problem ... re-flow / replace U7 (TL431)
I have the same model. It just died on me. Zero amp on the display and it doesn't charge anything. I've opened it up and I don't see any components that are blown. Do you know what it might be the problem?
Good Day sir, i have same model but the issue is when i increase the output it doesn't increase more than 8.9 volts. Do you have any idea on what could possibly the fault? I already reaplaced the multi turn pot. But the problem stays.
hey brother i need your help, i have this power supply but when i turn it on and turn the current knob the green light does not come on so i can't adjust anything, i can't figure out how to get it working
Brilliant work. My PSU died when I was using it to drive an arcing spark generator (don't ask), but I found the circuit diagram for the same PSU but with a less max ampage. So the area that you worked on failed. One of the large MOSFET's went short, 2 of the surface mount transistors that you pointed to went short, and so did ones of the schotky diodes next to those surface mount transistors - I've replaced all those . The PSU currently pulses the display showing 000 when switched on. Now I'm wondering if the switching controller you pointed to has died? I think this is the circuit diagram I had (thought you might find it interesting or useful) = www.eevblog.com/forum/repair/permanent-cc-and-no-output-(longwei-k3010d)/
the same thing to me with the short, but I also changed the U5 TNY820 AND THE PC817B PHOTO COUPLER, IT HAS RESUMED WORKING, PULSA X WHICH IS PROTECTED THROUGH U4 SG3525.
andy parry oops I got confused, the component I replaced was U4 Sg3525 and that's it.
@@diegochiarenza1456 I tried that and it didn't help much. I suspect one of the amp has gone.
@@andyspoo2 mi dispiace.....
Can you please send a link to the circuit you found? Thanks
what is the reason for removing the noise? i am thinking of powering a radio transceiver that requires "12v", i have not done this yet, but would this noise come through the radio some how?
Hi, the switching noise if high frequency and have a broad bandwidth, it will radiate out into everything. If you try to measure something on a scope it gets very annoying because you don't know if the noise is from the device you're working on or it's from the power supply and it will make it impossible to make any kind of sensitive measurements. It will also create radio interference, as in all kinds of radio equipment will be affected, even if they're not powered from the supply, it's enough the supply is just switched on. Generally switching noise is very undesirable, especially in any kind of electronics lab environment.
Cheers,
Jake
Good video bro. I have this power supply brought from aliexpress, but cant work, because protect led always on. :(
good vid. I'm new to this.. Why not just leave it as i, why do you try to reduce noise what is the benefit??
Thanks
Noise is bad for sensitive circuits, coms, medical, etc. Switching power supplies are inherently noisy. Some are really noisy. Linear PSU's are much cleaner. Generally, less noise is better.
@@sekboi I have that same PS. BG is sending a replacement, it was so noisy I could not test anything with my scope because of the noise this PS putout the trigger would just not lock.
Thanks for the info.
You're welcome, if you own this power supply it's definitely worth your time to make this little modification, cheers.
Hi sir, i have the same power supply and the problem is that cannot keep Amps stable, up - down. Could you tell me sir where to look or what to replace?
Hello, the output current depends on the current limit and what load you have on the power supply.
To set the current limit you need to set the voltage to something e.g. 5V then short the output and you can adjust the current to what limit you need.
Hope that helps,
Cheers,
Jake
@@TheStuffMade i understand what you said sir it's my mistake, i didn't said it right. What i m just saying is, after set the voltage that i need, and adjust the current as you said on the display, initially is e.g. 2 Amps after a second it is 2,5 Amps then 3, 2, 2,5. it is increasing and decreasing by it self with stable voltage. What could i do?
In my opinion, the worst feature of this PS is the current spike when used as constant current supply at low levels. The behaviour reminds that of a line output with a relevant capacitor in parallel. Did you check if your mod do not worsen the already bad spike?
I have another video testing the speed and performance of this power supply:
th-cam.com/video/RdT-xWC4h4k/w-d-xo.html
It's typical for switch mode power supplies to react slowly to changes because of the way they work, most linear power supplies have superior performance in that regard, but switch mode supplies are strong when it comes to high power and lower cost.
The noise fix will not improve the speed of the power supply.
Yes it will limit the noises because of decreasing dv/dt
That means the mosfets switch slower(fall time and rise time increased)
So the mosfets will be get hot and ...dead
By adding RC snubber and changing the output caps with better (like solid caps) the noises will limit
Excellent
is the switching noise as dangeorous as ripple for electronics? Thank you :)
Neither are particular dangerous to electronics as long as they are within reason, but you always want a bench power supply to be a low noise as possible otherwise it can affect the behavior of your circuit and make it very difficult to do any reliable measurements.
Cheers,
Jake
Not using for medical electronic repair or running medical equipment...as a bench supply your safe.
i have a minleaf 30v 10a power supply i havent got a scope to check it though well i have but its one of those mini DSO138 i dont think its capable of these tests but i could be wrong.
These little scopes are good up to 100-200kHz. They are useful for audio etc. but for these noise measurements, as it's very high frequency switching noise, you need a scope capable of 100-200Mhz, so about 1000x faster to get good measurements.
Cheers,
Jake
R-SPS3010 or longwei(K3010D) . what is the best ?
Great video, thanks! They should just incorporate your fix into the power supply.... one can hope.
Thanks, yeah it would be great if they'd pay more attention to noise.
What if you put just the capacitors without the resistor on the gate
Hello David, you need the resistors, I didn't just add them for fun. If you don't add the resistors and only the caps you'll put way too much load on the driving circuitry and it'll likely let some magic smoke out.
Cheers,
Jake
I have a source of this LongWei LW-K3010D it is falling unregulated voltage does not let it go over 9V then turn on the sera fan q can be the mosfet???
It's really hard to say without doing some measurements with a multimeter and an oscilloscope.
Cheers,
Jake
Had the same issue. Probably a later revision board which now has a TL413 (U7) as a ref voltage. Check pin 1 as I bet it no longer maintains a constant voltage.
Very good video
Thanks
if I replace 10N60 transistor for 20N60 transistor will have any problem to circuit? I will need change something else ?
It would probably still work, but why would you want to change the mosfets?
@@TheStuffMade about the transistor, I ask just for curiosity, and my power supply is the same as the video, but is not changing voltage, just still at 0, I try replace PWM (SG3525A), and CI 084C, and LM2904, I replace the transistors, check de potentiometer BUT NOTHING!! I’m really upset, some idea? thanks for attention
Hi how does the Korad ka3005d vs LW-K3010D with mod compare in noice level ? i would like to compare Vpp and Vmax. Since i can get the K3010D for 56 dolars and the Korad for 85 dolars, both with shipping. Since for my understanding the Fix, does reduce alot the Vpp.
The Korad is the better power supply even after the fix. If I could have only one of these power supplies, I'd get the Korad. However the Longwei is great to have when you need a lot of power as it's capable of delivering more than 300W output. Make sure shipping is included when you compare the cost as the Korad is a linear power supply with a large transformer, so it's heavy and often cost a lot to ship.
Hope this helps,
Cheers, Jake
@@TheStuffMade thanks, the psu comes from spain, i live in portugal, with will be a total of 71 euros, so its a nice price.
@@tdomingues89 That's a very good price, maybe get 2 of them if you work on audio circuits.
Cheers,
Jake
Hi, I have a doubs of the capacitor you're using is a 33 o 47uF and what voltage thanks?
Hi Marco, it's 47nF not uF, use a film capacitor. The capacitor doesn't see any high voltage so e.g. a 50V rated capacitor is fine, or use one of the green ones I use, they are very common and I believe they are rated at around 100V.
Cheers,
Jake
@@TheStuffMade perfect thanks
@@TheStuffMade Would a 100nF work(yes, it will, but will it matter??) Have a few at hand!