Fixing the LongWei LW-K3010D Bench Supply Noise Issue

hello in this video I want to have another look at the long wise which more powerful so specifically one I want to have a look at is see if we can do anything about the noise issue the noise can get quite offensive when you're measuring things on the scope we can try and have a look here what it looks like so what we're looking at here so I have power supply hooked up here just set five wall outputs we have some kind of standard at baseline and then I hope to have a small load here's a piece attorney it's darling we're loading it up with 2.4 to MCM and I just have the scope attached so the scope is the test directly to the output and if we try to zoom in a little bit on the scope we can capture it here we kind of have two different kinds of noise here one the spikes here that is switching noise that is really the one that annoys me the most and then we have these smaller peaks and dips here so that is more the ripple of the power supply so I would like to see if I can do anything to get rid of at least low what is switching noise here because it is it is quite annoying so this pathway does actually also produce another type of noise on the higher load Christ in here if I up the voltage here see it's not generating this kind of sine wave pattern and but that's not really noise that's motive regulations I think it's just too slow regulating so it's kind of overcompensating every time it adjust the output so it will just too much and then it will come and say again just two months in and then you get this kind of sine wave pattern however I don't think there's much we can do about that and to say we did ripple I don't think that's much we can do about that but I want to have a go at the switching noise and we can do something about that I've taken the cover off the power supply here so I want to see if we can find the exact source of where the high frequency switching noise is generated so I don't have any fancy probe for this but you should just take a normal scope probe just take the tip off here and then take the ground wire let's twist it around one time and just connect it here short it out so that will give us a probe and as we've helped us locate exactly where the noise is coming from so I'm switching on D power supply now I have can already see on the scope that it's picking up the noise so if we start over here it's kind of moving it along here it's getting more and more and more and more and more and right here it's at the maximum I think yeah so this high frequency noise really gets radiated out of the power supply this is an easy way to locate where the noise is coming from so you'll have a good idea where to that's a look so the thing is with switching noise when you want to do something about it not really the best thing you can do is to eliminate it at the source trying to get rid of the swishing noise later on in the circuit and might not help anything because it's going to radiate out into everything and we can see this area here it corresponds with two or devices we have one heatsink on the other side so I will assume those are switching transistors so we could try have a closer look and see if we can figure out exactly how they're doing it so let's have a closer look at bored inside here so we can kind of see the primary high voltage side is over here well I think it goes from here there's an opto coupler down here yes it goes there you can see board cut out and they have some kind of current sends here and splits here with another transformer and then this is the main transformer and then we probably go working and so everything down here is high Walters over here they see a low voltage secondary so really like to see what these two devices are I kind of guessed they're gonna be two MOSFETs but it's really hard to see because they're kind of totally obscured by the transformer here and this device over here what is that so I managed to take a couple of pictures here not using some spotlights and we can see the devices here so this is a Fairchild and say 10 and 60 something and the old device over here in the MBR 3,200 C and the device on the other side of the transformer is also a fairchild device also 10 in 60 it looks like I see here so those two were the same that counter indicates it's a push-pull thing we got going on and I found the datasheet here so do you two transistors are 600 wall M channel MOSFET that is clearly set up is some kind of push-pull configuration on the primary side and I got another close-up picture here so we can see D switching controllers and SG 35:26 AP and we have the datasheet here for these with no controller and since it's located on the secondary low-voltage side I assume this is kind of the configuration they're using okay so these two over here are two MOSFETs and channel MOSFETs this is our main transformer here and down here this is the switch mode controller and it appears this transformer here on preview is this transformer used to to transmit the gate control signals over here from the primary side to the secondary side and we can see there are a few transistors and diodes and resistors down here I'm pretty sure that has something to do with driving the gates on these two MOSFETs here so I think the reason it's so noisy is because they're driving the gate quite hard which is good for efficiency and it's it will keep the MOSFET cold it's not going to be so much power dissipation in them however the downside is we get some pretty bad switching noise so I think when our attempt is to see if we can slow down the gates a little bit of course downside is we will be dissipating a bit more power they must waves but I don't think it's going to be a huge issue okay so I think I want to attempt at a resistor on the gate that should hopefully slow down the MOSFET switching a little bit so what we have here is the gate is over here then we have drain that we have source here and same over here gate drain source so I had a quick look but unfortunately it doesn't look like there's any way I can get a resistor in here without making bit out cording a trace and adding a resistor so I'll try one side first here and see if it makes any difference just gonna cut this Tracy I had a small resistor in there okay so I think I want to try everyday 10 or my sister first oh well let's take a Lebanon because it's less likely I'm gonna use a lemon or in the future it doesn't really matter okay I'm sort of into a sister how we could also try put a capacitor between the gate and the source if this resistor is not enough all right let's try power up and see what happens I changed my mind let's do both sides before we test it because it's gonna be hard to tell how well it's working if we only have one side and it's out of balance [Music] and here hold it up again so it might be a tiny bit better but it's certainly not good enough to call it a success so I think we'll have to do some more medications next I want to try add a small capacitor so I have DS this is a 33 nano farad so I think that's a good starting point so let's when I get a couple of those in so those are going to go between the gate and the source so I'll cut your legs a bit and I've bend it a little bit into shape because we don't want it to get too close to the primary side over here so let's try again and here we have connected again with D capacitors and now it's significantly better and have much much much less noise however we might be able to do even better I think I'll try opti resistors 220 or me set up the eleven on we have in now [Music] [Music] okay so it's looking pretty good now I'm fairly happy with this result here it is looking a lot better than it did before yeah the switching noise is definitely much much much lower and now we're getting down to such low noise that's actually the noise just in the environment here it's probably part of what we're seeing here doing these tests I've been running the scope in full bandwidth mode so that means to be almost 200 megahertz bandwidth here and using x10 on my probes here so I can't why sweets off the power supply and see what difference it makes yeah so you see what the power supply off we have nearly as much noise and just a quick little comparison here however it's not job done yet you still have to make sure the MOSFETs are not running too hot now because they will be dissipating a little bit more power or maybe a lot more power so we need to get a thermal camera out and meter load it up really hard and make sure it's still 100% stable so here the thermal camera now happy power supply loaded up with about 240 watt load just to see how hot it gets especially the MOSFETs however you can see the Noda one yeah it's getting pretty hot these are 100 watt resistors on a very last instinct eight pieces of 100 watt resistors and if we look at the powers upon you can see the bridge rectifiers down here well they can obscure thing about 60-something degrees and the semester here the input probably about the same and have a main transformer sitting less than 50 degrees already the MOSFETs over here not getting hot at all this is the diode on the output this choke and the output filter and not on show going here so really nothing looks like it's getting really hot except my local sister Sonia they are getting really hot so it looks good I think I could even go a little bit more aggressive with the capacitor on the gauge here since the MOSFETs was still running quite cold I decided to do one more modification so I change the capacitor to 47 nano farad and EEMA sister's 30 or so these are going to be the final values then I did a full load test again and you can see in thermal image here so the mass person just starting to heat up a little bit now almost nothing so let's do a final noise comparison test so here we haven't loaded up again if I want to point for 3m and this is still with full bandwidth so nearly 200 megahertz bandwidth here annika CDs switching noise is barely reasonable and let's just do one final comparison with the first measurement I did so it's a very significant difference handless to CD comparison with switch with the power supply off there we go that's the power supply off power supply on and loaded up so there's a nearly no difference it is very limited switching noise we have now I'm quite happy with this result here and the load test went fine 250 watt load for 5-10 minutes no problem and just a quick look here this is what a noise looks like you're on a scope with 20 megahertz bandwidth limiting this is how you usually measure power supplies so that looks quite nice so that's it for this video here our rate this is a huge success I hope you enjoyed the video if you did give a thumbs up subscribe if you haven't already subscribed and leave a comment below thanks for watching and I'll see you next time bye bye you

Channel: TheStuffMade

Views: 20,936

Rating: 4.9614973 out of 5

Keywords: Repair, LongWei LW-3010D, LW-3010D, Electronics, DIY, PSU Mod, Switchmode, Switching Noise, LW-K3010D, Long Wei

Id: WytDROmjWKQ

Length: 16min 53sec (1013 seconds)

Published: Wed Jan 22 2020