Falcon PSU recap - 28uS spike

Repairs & upgrades.
User avatar
exxos
Site Admin
Site Admin
Posts: 18644
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: Falcon PSU recap - 28uS spike

Post by exxos »

This is what I was meaning with spikes relating to the switching frequency. I think your spikes are not relating to this as far as I can tell..

It's why you need to measure similar to see what's really going on.

Capture.PNG
Capture.PNG (290.78 KiB) Viewed 160 times
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
Steve
Posts: 2007
Joined: Fri Sep 15, 2017 11:49 am

Re: Falcon PSU recap - 28uS spike

Post by Steve »

But he said he's got two recapped PSUs, if he's testing them both in the same environment, with the same equipment, then if one displays this spike but not the other - then surely that rules out environment?
User avatar
exxos
Site Admin
Site Admin
Posts: 18644
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: Falcon PSU recap - 28uS spike

Post by exxos »

Steve wrote: Wed Jun 22, 2022 11:47 pm But he said he's got two recapped PSUs, if he's testing them both in the same environment, with the same equipment, then if one displays this spike but not the other - then surely that rules out environment?
Don't know..maybe he had a laptop or PC on near it in one test and not another. Maybe he had LED lighting on in one test and not another.. He could have changed rooms between tests. The whole enviroment is unknown. Its why I just suggested a different room rather than getting into it all. Its a simple test to rule it out.

I guess a retest of the other PSU might be in order just to make sure aswell..

Currently more will be known if he can see the ripple and noise better. If he's not seeing noise spikes over the ripple, then I don't see how it can be the PSU causing the issue.
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
foft
Posts: 65
Joined: Mon Mar 28, 2022 12:20 pm

Re: Falcon PSU recap - 28uS spike

Post by foft »

With the scale at 10mV and at 2uS time base I do see a sine wave, of sorts. Its a right mess:

ScreenImg.png
ScreenImg.png (25.74 KiB) Viewed 127 times

Zooming out to 5uS you can see a lower frequency repetition here:

ScreenImg-2.png
ScreenImg-2.png (28.78 KiB) Viewed 127 times

And then at 50us:

ScreenImg-3.png
ScreenImg-3.png (30.31 KiB) Viewed 127 times

Each slow oscillation is about ~30us, so ~34KHz. Then the higher frequency oscillation (top image) is ~700KHz.
foft
Posts: 65
Joined: Mon Mar 28, 2022 12:20 pm

Re: Falcon PSU recap - 28uS spike

Post by foft »

Anyway I broke out the other PSU and ... see the exact same thing! I didn't notice it before, but I wasn't looking in such detail I think (or perhaps scope was set to the 20MHz bandwidth limit).

ScreenImg-4.png
ScreenImg-4.png (34.36 KiB) Viewed 122 times
ScreenImg-5.png
ScreenImg-5.png (29.96 KiB) Viewed 122 times

Then the much larger one occasionally:

ScreenImg-6.png
ScreenImg-6.png (24.72 KiB) Viewed 122 times
ScreenImg-7.png
ScreenImg-7.png (26.16 KiB) Viewed 122 times

So I guess this is 'normal', either a feature of the PSU or a feature of my room... The fact its 34KHz and that matches the general oscillation makes me think its the PSU.
User avatar
exxos
Site Admin
Site Admin
Posts: 18644
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: Falcon PSU recap - 28uS spike

Post by exxos »

foft wrote: Thu Jun 23, 2022 8:49 am Anyway I broke out the other PSU and ... see the exact same thing! I didn't notice it before, but I wasn't looking in such detail I think (or perhaps scope was set to the 20MHz bandwidth limit).
Thanks for re-testing. It saves some confusion should anyone else read this thread in the future :)
foft wrote: Thu Jun 23, 2022 8:49 am So I guess this is 'normal', either a feature of the PSU or a feature of my room... The fact its 34KHz and that matches the general oscillation makes me think its the PSU.
What threw me is like in your last image where the spikes are not consistent to the ripple and seem the same amplitude. So this would imply its not the PSU..BUT. Now you zoomed in the ripple you can now see the spikes as they would normally show in sync with the ripple and more random spikes.

If you measure the p-p of your capture, I assume the ripply would be something like 50mV, then the spikes (noise) would probably be 200mV ? That would pretty much be normal with these old supplies. Though odd the re-cap hasn't squashed the spikes totally like in my tests. But indeed measuring such low voltages isn't simple. Like I mentioned earlier, if the scope probe or gnd wire is anywhere near the PSU then you can introduce noise that way. You see it on the scope, but it's not really there. When your down to such low voltages (<0.05v) then any electrical activity in your room can cause issues. LED lighting is a nightmare. Dave jones EEVblog even did a video on that. So you normally would do tests in a "silent room" and read pages of manufactures app-notes on scope grounding techniques and the rest of it. It gets rather in-depth. Spikes are also down to "cheap inductors" used on PSU's. There is only so much you can do to "silence them".

TL;DR
If you only spiking like 200mV then thats like 5.0V + 0.2V spikes. As long as your not spiking several volts then I wouldn't worry about it to much. It's also why I suggest replacing the 4,700uF cap on the motherboard with one of my kits as the multiple caps will greatly reduce the spikes before it gets to any silicon anyway.
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
foft
Posts: 65
Joined: Mon Mar 28, 2022 12:20 pm

Re: Falcon PSU recap - 28uS spike

Post by foft »

Many thanks for the help. I'm still not 100% sure if this is a scope artefact or just very short lived high frequency noise, still since the two PSUs match I decided to sign them both off for now (i.e. use them!). I might add some small fast caps on the output later.
Post Reply

Return to “PSU (power supplies)”