How to quickly & easily fix coil-whine(coil choke noise)

[bobogdan625]

Hello guys, I tried the glue method(used a ethyl 2-CA glue, ethyl is not that strong of a solvent so it does not affect the PCB) and had no improvement. I left the glue to cure overnight (about 12-14h) and this morning I added a bit more glue, fired up RDR2 to test it and the whining still persists.

I can attach a picture if you guys want.

What should I do next? I tried this on a Powercolor Fighter 6700XT. I heard that using 2x8 pins cables instead of a split 1x8 like I have may drown the whine a little. What do you guys think?

Undervolting the AMD cards is an absolute nightmare, in fact, AMD cards cannot be undervolted, you push the voltage slider down and the frequency goes down with it, you cannot undervolt without losing some performance, plus the voltage does not stay locked like it does with nVidia. I had an EVGA SC2 GTX1080 and that thing was a dream, super easy undervolting and did not lost any performance, only the power draw went down, the performance stood the same, like it should be.

 

[lexluthermiester]

Hello guys, I tried the glue method(used a ethyl 2-CA glue, ethyl is not that strong of a solvent so it does not affect the PCB) and had no improvement. I left the glue to cure overnight (about 12-14h) and this morning I added a bit more glue, fired up RDR2 to test it and the whining still persists.

I can attach a picture if you guys want.

What should I do next? I tried this on a Powercolor Fighter 6700XT. I heard that using 2x8 pins cables instead of a split 1x8 like I have may drown the whine a little. What do you guys think?

You may have missed the culprit chokes or maybe something else is at play. Take a few photos of the naked PCB and edit in indicators which point out what parts you tried to seal. Also show the glue you used, you might have used the wrong type. IIRC, Ethyl-2 isn't the type that will work well for this kind of thing. I'll have to look it up later and reply here. I'm not currently where I can sit down and deep dive this problem. Later today very likely.

Just an FYI, CA can be a solvent to certain materials. PCBs and any electronics parts mounted on them are not on that list.

Edit: corrected for a word left out.

 

[bobogdan625]

Thank you for your answer, I will comeback tommorow with lots of pictures. I found out after I made that post that I have 2 more inductors under the backplate, also glued them and now I am waiting for it to cure.

 

[lexluthermiester]

Thank you for your answer, I will comeback tommorow with lots of pictures. I found out after I made that post that I have 2 more inductors under the backplate, also glued them and now I am waiting for it to cure.

Ok, I'll be here. Will try to help you however possible.

 

[bobogdan625]

Here they are. I tried using a toothpick( tho my glued had a really small nozzle) and some inductors are still not completely surrounded with glue, they still suck the glue inside. I don't know what to do anymore, I have this gpu since July 2022 and I tried multiple things and this method is the closest I could get to fix the whine, I tested in multiple instances like letting it idle for a while while recording(for comparison, and it seems the idle whine is almost gone) and in demanding games like RDR2, instance when the whine is still there.

 

[lexluthermiester]

Here they are. I tried using a toothpick( tho my glued had a really small nozzle) and some inductors are still not completely surrounded with glue, they still suck the glue inside. I don't know what to do anymore, I have this gpu since July 2022 and I tried multiple things and this method is the closest I could get to fix the whine, I tested in multiple instances like letting it idle for a while while recording(for comparison, and it seems the idle whine is almost gone) and in demanding games like RDR2, instance when the whine is still there.
View attachment 362378View attachment 362379View attachment 362380View attachment 362381View attachment 362382View attachment 362383View attachment 362384View attachment 362385View attachment 362386

You might have used a bit much, but I can't find anything you missed. You did a decent enough job of it that if the problem was the chokes on the card there should have been some level of result. Maybe mobo or PSU? Or you might just have a card that is noisy beyond what can be suppressed. Sadly, it goes that way sometimes.

 

[Bill_Bright]

It is possible the vibration causing the whining is happening inside the "can" of the offending component and I don't know of anyway to fix that other than to replace that component.

I am not sure what I am looking at when I look at your images. What I do see appears to be a lot of dust. Look at all the dust on the finger contacts on the right side of the 2nd image from the bottom. The first 4 images look pretty clean. The rest do not - at least from here. If me, I would have made sure those boards were sparkling clean before applying any sort of glue or TIM or anything else.

 

[bobogdan625]

You might have used a bit much, but I can't find anything you missed. You did a decent enough job of it that if the problem was the chokes on the card there should have been some level of result. Maybe mobo or PSU? Or you might just have a card that is noisy beyond what can be suppressed. Sadly, it goes that way sometimes.

Well, I tried a Corsair RM650X 2years ago and it still produced whine(not even a slight improvement), I don't suspect the MOBO(B650 Gaming X) cuz the whine is correlated to the GPU core frequency. Maybe this GPU is indeed cursed. Thinking of buying a nvidia gpu next time, waiting for 5000 series.

It is possible the vibration causing the whining is happening inside the "can" of the offending component and I don't know of anyway to fix that other than to replace that component.

I am not sure what I am looking at when I look at your images. What I do see appears to be a lot of dust. Look at all the dust on the finger contacts on the right side of the 2nd image from the bottom. The first 4 images look pretty clean. The rest do not - at least from here. If me, I would have made sure those boards were sparkling clean before applying any sort of glue or TIM or anything else.

I used multiple Q-tips (or cotton buds for my fellow UK readers) for cleaning. Those contacts covered in dust were HDMI/DP ports and they don't really care whine-wise.

 

[lexluthermiester]

Well, I tried a Corsair RM650X 2years ago and it still produced whine(not even a slight improvement), I don't suspect the MOBO(B650 Gaming X) cuz the whine is correlated to the GPU core frequency. Maybe this GPU is indeed cursed. Thinking of buying a nvidia gpu next time, waiting for 5000 series.

Situations where no improvement takes place are you usually caused when the coil in question is vibrating against the choke coil enclosure wall. The only way to solve it then is to either desolder the problem chokes and seal them directly, or to replace the chokes themselves with units that have identical ratings. Either way, it is a much more involved process. Not for the inexperienced or faint of heart, but not greatly risky either if one is careful.

 

[KLiKzg]

Guys, do note that: harder the cured glue used, less vibrations it can absorb. So it is better to use glue which is not so hard.

 

[bobogdan625]

Situations where no improvement takes place are you usually caused when the coil in question is vibrating against the choke coil enclosure wall. The only way to solve it then is to either desolder the problem chokes and seal them directly, or to replace the chokes themselves with units that have identical ratings. Either way, it is a much more involved process. Not for the inexperienced or faint of heart, but not greatly risky either if one is careful.

I have 0 soldering skills . Where do you even get those chokes that have identical ratings?

 

[Bill_Bright]

Guys, do note that: harder the cured glue used, less vibrations it can absorb. So it is better to use glue which is not so hard.

Not necessarily. If talking silicone adhesives, then being flexible, and if properly applied, it can act like a cushion, similar to a pillow between the headboard and the wall.

But a very hard, solid adhesive, if properly applied, can stop the vibration completely.

 

[Solaris17]

I have 0 soldering skills . Where do you even get those chokes that have identical ratings?

This is where I get everything when fixing something or working on custom stuff.

https://www.mouser.com/

they pretty much carry every SMD known to man.

 

[Shrek]

Except the capacitors I want are sold in lots of 200

 

[Bill_Bright]

I have 0 soldering skills

As you note, soldering is a "skill" and as such, they need to be practiced regularly and "honed". For this reason, for those who solder (professionals and novices alike), instead of tossing those old motherboards, graphics cards, and other electronics into the recycle bin, use them to practice (and touch-up) your soldering (and desoldering) skills by removing and replacing components on them - especially before working on your good parts. I occasionally do this to verify my soldering pencil tip temperature is right.

 

[lexluthermiester]

Guys, do note that: harder the cured glue used, less vibrations it can absorb. So it is better to use glue which is not so hard.

That's not true at all. Super glue\CA glues are not employed to absorb vibrations at all in this use case situation. The use case in this kind of application is to seal the noisy choke so the noise can not come out of the choke housing or is greatly suppressed and hopefully glue the coil itself in place so it doesn't produce noise to begin with.

I have 0 soldering skills . Where do you even get those chokes that have identical ratings?

It depends on where you are in the world.

This is where I get everything when fixing something or working on custom stuff.

https://www.mouser.com/

they pretty much carry every SMD known to man.

Here stateside, Mouser is the best!

Except the capacitors I want are sold in lots of 200

Well, you'll have spares for a while.

 

[KLiKzg]

That's not true at all. Super glue\CA glues are not employed to absorb vibrations at all in this use case situation. The use case in this kind of application is to seal the noisy choke so the noise can not come out of the choke housing or is greatly suppressed and hopefully glue the coil itself in place so it doesn't produce noise to begin with.

I do like you, but this is a common physics. Why do you think than "silent rooms" are from material that absorbs audio (waves)?

So you can't audio isolate the house with hard wood - like oak - on a wall or by putting stone wall.
But you can do it with softer (& more prone to elongation) wood, like pine for example. Or some softer material.

Same way: if you use the softer hardening glues (which are also silicon based) or similar. That is also why automotive industry (in which I was working) was using such glues. So my suggestions come from experience & practical work. Take it or leave it, your choice!


BTW, my GPUs are working 24/7 on almost 100%...so I do not have these kind of issues!

 

[lexluthermiester]

I do like you, but this is a common physics.

Sure it is. You're misunderstanding the way these glues are being used. See below.

Why do you think than "silent rooms" are from material that absorbs audio (waves)?

Those types of materials are NOT CA glues. Once again, SuperGlue/CA glues being used in THIS type of application are not to provide sound insulation. These glues are being used to seal the choke housing to the PCB, thus sealing the noise into the housing and lock in place the coils themselves in an optimal situation. That's it. There is no insulative effect at play. It's not intended or inferred. At all. Nor is such taking place.

 

[KLiKzg]

Sure it is. You're misunderstanding the way these glues are being used. See below.

Those types of materials are NOT CA glues. Once again, SuperGlue/CA glues being used in THIS type of application are not to provide sound insulation. These glues are being used to seal the choke housing to the PCB, thus sealing the noise into the housing and lock in place the coils themselves in an optimal situation. That's it. There is no insulative effect at play. It's not intended or inferred. At all. Nor is such taking place.

& again, saying that is wrong.

Why?
As the noise insulated that way with "hard glue", will transfer to the:

• housing of the coil, which is from harder (usually plastic) material & will continue to vibrate.
• to the PCB itself, which will transfer the noise to the: GPU casing with fans, motherboard through PCIe & to the case itself. Note, than any irregularities in soldering might give you problems with vibrating PCB & might lead to going your GPU card dead or burnt.

So you have not got rid of the vibration, at the source. Just encased it or transferred it to larger device in hope the housing or the larger GPU PCB will keep it unnoticeable.
To me, that is not the best way to deal with vibrations.


Or take it this way:

• your way of taking rid of vibrations from a choke is like having a fan that rattles & you fix it with screws to a case
• my way of taking rid of the vibrations from a choke is like having a fan that rattles & put it on the anti-vibration silicone inserts

Which fan would give out more noise?

 

[lexluthermiester]

& again, saying that is wrong.

Why?
As the noise insulated that way with "hard glue", will transfer to the:

• housing of the coil, which is from harder (usually plastic) material & will continue to vibrate.
• to the PCB itself, which will transfer the noise to the: GPU casing with fans, motherboard through PCIe & to the case itself. Note, than any irregularities in soldering might give you problems with vibrating PCB & might lead to going your GPU card dead or burnt.

So you have not got rid of the vibration, at the source. Just encased it or transferred it to larger device in hope the housing or the larger GPU PCB will keep it unnoticeable.
To me, that is not the best way to deal with vibrations.


Or take it this way:

• your way of taking rid of vibrations from a choke is like having a fan that rattles & you fix it with screws to a case
• my way of taking rid of the vibrations from a choke is like having a fan that rattles & put it on the anti-vibration silicone inserts

Which fan would give out more noise?

This is going in circles and I'm not going to argue further.

 

[Bill_Bright]

So you can't audio isolate the house with hard wood - like oak - on a wall or by putting stone wall.

Ummm, of course you can. They would have be very thick but it works. Note in the better hotels, the walls and especially floors between rooms do a very good job of blocking sound. Of course, if your neighbor bangs on the wall or jumps up and down on the floor, you'll hear it. But if those barriers are thick enough, you won't. I note hardened clay makes a great sound barrier.

 

[bobogdan625]

Hello again, sorry for not responding for a few days. I came back to this thread after struggling to dampen the sound of my case. I used an Apple Watch Series 9 (which has a built-in decibel meter that's said to be pretty accurate) to measure the decibels before and after, and the results were quite surprising. Before, the readings were 42-44 dB, and now they've dropped to as low as 38, with a maximum of 41. Most of the time, the average is around 40 dB. Note that the tests were done in gaming(RDR2 1440p High-ultra settings). In idle, it sits around 37-38dB, sometimes even drops to 36 and that's with Arctic P12, I can't wait to get my hands on a P12 Max. Since dB use logarithmic scale, after my calculations, the improvement is roughly 37 to 60%(perceived by the human ear). Well worth it.

I used StP Biplast with a thickness of 10mm (it costs about $30-35 in my country) and some StP anti-squeak(1-2$) tape for sealing. My GPU temperature rose by 1°C, but the tradeoff is definitely worth it. I can upload some videos showing the before and after if you guys want.

 

[lexluthermiester]

Hello again, sorry for not responding for a few days. I came back to this thread after struggling to dampen the sound of my case. I used an Apple Watch Series 9 (which has a built-in decibel meter that's said to be pretty accurate) to measure the decibels before and after, and the results were quite surprising. Before, the readings were 42-44 dB, and now they've dropped to as low as 38, with a maximum of 41. Most of the time, the average is around 40 dB. Note that the tests were done in gaming(RDR2 1440p High-ultra settings). In idle, it sits around 37-38dB, sometimes even drops to 36 and that's with Arctic P12, I can't wait to get my hands on a P12 Max. Since dB use logarithmic scale, after my calculations, the improvement is roughly 37 to 60%(perceived by the human ear). Well worth it.

I used StP Biplast with a thickness of 10mm (it costs about $30-35 in my country) and some StP anti-squeak(1-2$) tape for sealing. My GPU temperature rose by 1°C, but the tradeoff is definitely worth it. I can upload some videos showing the before and after if you guys want.

Yeah, chime in! What you're suggesting seems very interesting!

 

[KLiKzg]

Vibration is NOT noise!

Noise is sound, typically unpleasant or unwanted sound. Sound is caused by vibrations, but that does NOT imply that vibration is sound. Lots of things vibrate without making any sound.


Yes, the source of the vibration was the coil, but the source of the sound came from something else. I had hoped my analogy of the noisy, rattling muffler caused by the vibrating car engine (which you even quoted) would illustrate that. Sorry it apparently went right over your head.

Well, this is wrong...almost every vibration is making a sin wave in the medium (air)...so it makes a sound.
Why? There is a lower limit of molecule frequencies that some sin wave can start & upper.

Just because our ear can't hear sound, that does not mean it is not there...it is still vibrating! Just outside of our perception.
So maybe some GPU are still making a sound, but it is only audible to bats!

 

[lexluthermiester]

Just an FYI, CA can be a solvent to certain materials. PCBs and any electronics parts mounted on them are on that list.

Just re-read this and need to correct;

"Just an FYI, CA can be a solvent to certain materials. PCBs and any electronics parts mounted on them are not on that list."

Well, this is wrong...almost every vibration is making a sin wave in the medium (air)...so it makes a sound.
Why? There is a lower limit of molecule frequencies that some sin wave can start & upper.

Just because our ear can't hear sound, that does not mean it is not there...it is still vibrating! Just outside of our perception.
So maybe some GPU are still making a sound, but it is only audible to bats!

Ok, enough with the running in circles. Please stop with this pointless line of discussion. Thank You.