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

[Frag_Maniac]

10ml,5ml,3ml Syringe & Dispene Blunt Tip Nozzle For Glue Please select size | eBay UK

We aspire to the highest of standards both in our products and service. Tip sizes include Dark Green 14 gauge, Brown 15 gauge, Black 16 gauge, Jade Green 18 gauge, Light pink 20 gauge, Purple 21 gauge, Blue 22 gauge, Orange 23 gauge, Red 25 gauge, Clear 27 gauge.

www.ebay.co.uk

Thanks, is there any documentation or stampings that indicates what gauge range the needles are? I've read penetrating epoxy can be pushed through a 20 gauge needle. I may have to experiment with some denatured alcohol I have to thin it out a bit.

Never mind, oddly enough your embedded post shows the gauges, but I didn't find them on the product page.

 

[Arcsaber]

Thanks, is there any documentation or stampings that indicates what gauge range the needles are? I've read penetrating epoxy can be pushed through a 20 gauge needle. I may have to experiment with some denatured alcohol I have to thin it out a bit.

Never mind, oddly enough your embedded post shows the gauges, but I didn't find them on the product page.

"Tip sizes include:

Dark Green 14 gauge, Brown 15 gauge, Black 16 gauge, Jade Green 18 gauge, Light pink 20 gauge, Purple 21 gauge, Blue 22 gauge, Orange 23 gauge, Red 25 gauge, Clear 27 gauge"

But I couldn't find the 27g in my set...

 

[Frag_Maniac]

But I couldn't find the 27g in my set...

Yeah and the Clear one doesn't show in the pic either. I was thinking a Blue 22 Gauge needle might be small enough for my purposes, but I'd probably use a metal file on either side of the tip to give it a wedge shape, which would make it easier to slip in between the the brush bristles.

I may be overcomplicating it though, I've seen a guy who reinforces the bristles on the brushes he uses to apply epoxy by just dabbing the outside edges of the bristles near the ferrule, then wiggling the bristles to work it in. This would also minimize drips as you can use a thicker epoxy.

I keep envisioning myself getting it all over and ruining the bristles by making them clump together too much. I'm not the greatest with glue, kind of like the guy who always steps in the chewing gum someone tosses on the sidewalk and can't get it out of his shoe!

 

[taka]

Hi

@lexluthermiester Thank you for this great thread!

I have an Asus TUF RTX 2080 Super with coil whine.
The coils in this card look like this:

Anyone had success with those types? They look like closed casings.

Otherwise I might just have to try it.

I have a Asus TUF RTX 3070 with the same SAP II chokes and it was singing in tandem with the Corsair PSU (8 years old). I changed the PSU with another CWT branded Thermaltake GF1 750 and no more coil whine

 

[Roflewaffle47]

hey guys, first time attempting something like this. can anyone tell me which of these parts could be likely causes noise makers?I want to try OP's method. Here is the board.
This is the Intel arc a770 by Asrock "phantom Gaming" edition
View attachment ASRock-Arc-A770-Phantom-Gaming-11.webp

 

[Shrek]

I've never tried that. Does cooling them preserve them?

Would seem so.

 

[lexluthermiester]

I have a Asus TUF RTX 3070 with the same SAP II chokes and it was singing in tandem with the Corsair PSU (8 years old). I changed the PSU with another CWT branded Thermaltake GF1 750 and no more coil whine

This makes me wonder: Is there a modulation resonance effect going on with certain combinations of components?

Would seem so.

Gonna have to try that.

 

[taka]

This makes me wonder: Is there a modulation resonance effect going on with certain combinations of components?

The first time i installed the RTX 3070 the noise was also induced in soundcard OP-AMP's and CPU power stage. PSU and GPU were the loudest. All components vibrating at the same frequency as the GPU whine. Like the signal returned on the 12v line, fact confirmed with an Osciloscope. The old PSU was a Corsair VS 550 and close to it's limmit, all system using like 390W. After the PSU was changed there is still a whine on the GPU chokes but only can hear it when i turn off all fans.

And BTW the signal (ripple) comes from the MOSFET and resonated into the chokes. If you remove the choke and listen the noise it's still there in the MOSFET case/package.

https://www.onsemi.com/pdf/datasheet/ncp302155-d.pdf

 

[Bill_Bright]

This makes me wonder: Is there a modulation resonance effect going on with certain combinations of components?

I would hope not. If there was, I would suspect a defect somewhere - if not in a component then in the design. And if design, it likely would be a problem for everyone with the same HW configuration.

Point being, any such resonance, feedback, or whatever it is should not happen. It should never occur in the first place, or if inevitable, then by design it should be isolated and/or thoroughly suppressed BEFORE it becomes an "interference" issue in another circuit.

 

[lexluthermiester]

Point being, any such resonance, feedback, or whatever it is should not happen.

Agreed, if all components are properly engineered it should never happen.

 

[Bill_Bright]

Agreed, if all components are properly engineered it should never happen.

And with no manufacturing defects or otherwise damaged after production.

 

[Simone]

I've never tried that. Does cooling them preserve them?

Glue sets when the moisture in the air reaches the solution.

Fridges remove moisture.

Another thing to do is keep desiccant bags, throw them in the oven at 200*C for an hour to dry them out, then keep them inside ziplock bags and toss the glue in with it. :3

 

[Bill_Bright]

Fridges remove moisture.

Ummm, not from sealed containers. The glue is [hopefully] in a sealed container. This means refrigerators would have no effect on the moisture content of the glue. Neither would desiccant bags.

If the tube, tub, or bottle of glue is not properly sealed, it does not matter where it is stored. Exposure to the air will soon dry/set the glue making it useless for future use.

 

[Roflewaffle47]

I gave it a try, didn't seem to work for me. maybe i didn't have enough?
I did wait the 24hrs

 

[xrobwx71]

STARBOND

Want to Save Time Filling Small Cracks, Voids, and Knots?

www.amazon.com

According to Amazon, I purchased this (one 2oz bottle of thin, one 2oz bottle of medium) on April 9, 2022. I've used them probably 6-7 times. I store it on a shelf inside my climate-controlled home. No fridge, no oven. I checked it at the time of my last post and it's in perfect condition.

The way the bottles are made, they re-seal like new.

 

[Bill_Bright]

I gave it a try, didn't seem to work for me. maybe i didn't have enough?
I did wait the 24hrs

Sorry that did not work. It is not a guaranteed fix. In fact, other than replacing the offending component, there is no guaranteed fix.

Assuming you had the correct offending component, even if the super glue is very thin, there's no assurance it could work its way into the affected area. Also, as I said a ~800 to 900 posts back, I usually use hot glue sticks because once set, super glue can become brittle. But the problem with hot glue is it is very VERY thick, even when melted so if the vibration is coming from buried layers of coils, it may not reach where it needs to go.

 

[lexluthermiester]

I gave it a try, didn't seem to work for me. maybe i didn't have enough?
I did wait the 24hrs

Take a few photo's as close up as you can and let's see what you have.

 

[Arcsaber]

I tried it, and while it didn't work for me, I will relate my experiences - it does not seem to do any harm, so if you're going to field strip your card anyway, you might as well:

1. I strongly recommend a syringe - it takes so little glue to fill each choke (less than 0.25 ml), that you need as much control as possible; my inductors had notches at either end, so I was able to get my needle under there:

As soon as you see glue weeping out from under the choke, stop - that's as much as it can hold.

2. Wait 24 hrs for the card to dry - superglue dries on contact with moisture, and it takes AGES for open patches to absorb it; if you put the card back together before it dries, it will continue off-gassing onto your heatsink or waterblock, and turn the metal dull and cloudy.

 

[lexluthermiester]

I tried it, and while it didn't work for me, I will relate my experiences - it does not seem to do any harm, so if you're going to field strip your card anyway, you might as well:

1. I strongly recommend a syringe - it takes so little glue to fill each choke (less than 0.25 ml), that you need as much control as possible; my inductors had notches at either end, so I was able to get my needle under there:
View attachment 351679
As soon as you see glue weeping out from under the choke, stop - that's as much as it can hold.
View attachment 351680
2. Wait 24 hrs for the card to dry - superglue dries on contact with moisture, and it takes AGES for open patches to absorb it; if you put the card back together before it dries, it will continue off-gassing onto your heatsink or waterblock, and turn the metal dull and cloudy.

Take a photo of the whole board. Let's see all of the other chokes. The main power rail chokes are not always the source of noise.

 

[Simone]

Ummm, not from sealed containers. The glue is [hopefully] in a sealed container. This means refrigerators would have no effect on the moisture content of the glue. Neither would desiccant bags.

If the tube, tub, or bottle of glue is not properly sealed, it does not matter where it is stored. Exposure to the air will soon dry/set the glue making it useless for future use.

Works for me, with shitty brands of cyanacrolyte at least in my experience. If I keep it in the fridge it stays good a lot longer than if I forget to put it in.

 

[Bill_Bright]

If I keep it in the fridge it stays good a lot longer than if I forget to put it in.

I would not know about that as I never dated the containers after opening them to actually keep track of elapsed times. All I know is, time passes at totally different rates for me. Some days seem to go by at normal speed, others seem to drag on. Yet the years keep flying by at lightspeed.

My point is, I would have no clue how long it has been since I opened the tube, or last used it - regardless where I stored it unless I put a date on it.

I guess I should start doing that.

 

[xrobwx71]

All I know is, time passes at totally different rates for me. Some days seem to go by at normal speed, others seem to drag on. Yet the years keep flying by at lightspeed.

Time is relative to the observer. (Some German guy said this)

 

[lexluthermiester]

Time is relative to the observer. (Some German guy said this)

That is theory, not yet proven fact.

 

[Bill_Bright]

True but then lots of theories, widely accepted as fact, have not been proven. The Big Bang Theory being one of them.

 

[Shrek]

There are no proofs in science; that is the regime of mathematics, but the evidence for a big bang is compelling.

However, this is the wrong forum, it being super glue and all.