I have attached 2200uF capacitors to all voltage rails in parallel with 0.1uF.
They are positioned as close to the scope as I can get.
After the capacitors I have clipped on ferrite beads from video projector power cables to all rails in the hope this may help.
I also reconnected loads to all voltage rails of my ATX PSU's meeting the Antec minimum load requirements.
Result:---
The scope behaviour has changed and seems LESS stable now but I can see the noise on the scope waveform at 2mV has decreased noticably.
I might try without the beads to see if they are making things worse.
Next up is isolating the grounds and then the big task of shortening all wires.
They are positioned as close to the scope as I can get.
After the capacitors I have clipped on ferrite beads from video projector power cables to all rails in the hope this may help.
I also reconnected loads to all voltage rails of my ATX PSU's meeting the Antec minimum load requirements.
Result:---
The scope behaviour has changed and seems LESS stable now but I can see the noise on the scope waveform at 2mV has decreased noticably.
I might try without the beads to see if they are making things worse.
Next up is isolating the grounds and then the big task of shortening all wires.
Sorry about the mixed results. I don't think the beads hurt anything. They're harmless unless you add them to RF circuits 
As I feared orginally, it's also possible something else is wrong that isn't power related. It's hard to say. If you can get your hands on another scope (school lab or whatever) you can hook the scope probes (ground and the tip) to where each rail enters the scope boards and see what the power looks like there--especially when the scope glitches/reboots. You can set it up to first trigger on any positive glitches, then negative (ground "bumps" can cause either polarity). You can also cross probe across the grounds. Just be careful with the 2nd scope's ground creating ground loops (or worse shorting out one of the supply rails back through the AC line ground).
As I feared orginally, it's also possible something else is wrong that isn't power related. It's hard to say. If you can get your hands on another scope (school lab or whatever) you can hook the scope probes (ground and the tip) to where each rail enters the scope boards and see what the power looks like there--especially when the scope glitches/reboots. You can set it up to first trigger on any positive glitches, then negative (ground "bumps" can cause either polarity). You can also cross probe across the grounds. Just be careful with the 2nd scope's ground creating ground loops (or worse shorting out one of the supply rails back through the AC line ground).
Id like to harm RF as much as I can in this case !
I was scoping the +5V supply this time and it is noisy, lots of hash around 100mV and large spikes of around 300/400mV peak.
Unfortunatly im in a 'between labs' situation and am stuck with my primitive test gear, i have a 2MHz Tek scope that i might be able to use but its terrible with triggering.
I really wish i had some low noise bench supplies to run it up on
It appears LeCroy did connect the grounds togeather on thier PSU despite keeping seperate wires back to it. I guess my star ground is ok then.
Also what do you think about adding those inductors I pictured two posts ago?
!I was scoping the +5V supply this time and it is noisy, lots of hash around 100mV and large spikes of around 300/400mV peak.
Unfortunatly im in a 'between labs' situation and am stuck with my primitive test gear, i have a 2MHz Tek scope that i might be able to use but its terrible with triggering.
I really wish i had some low noise bench supplies to run it up on
It appears LeCroy did connect the grounds togeather on thier PSU despite keeping seperate wires back to it. I guess my star ground is ok then.
Also what do you think about adding those inductors I pictured two posts ago?
Craig405 said:Id like to harm RF as much as I can in this case
I was scoping the +5V supply this time and it is noisy, lots of hash around 100mV and large spikes of around 300/400mV peak.
Unfortunatly im in a 'between labs' situation and am stuck with my primitive test gear, i have a 2MHz Tek scope that i might be able to use but its terrible with triggering.
I really wish i had some low noise bench supplies to run it up on
Also what do you think about adding those inductors I pictured two posts ago?
Tek even *made* a 2 Mhz scope??? Was that in 1933?
Noise of the levels you've observed on digital supplies is fairly normal. That's not to say it is normal for your scope, but most digital stuff doesn't care about about a few hundred mV of peak noise on a 5 volt supply. As I said, the critical stuff in your scope is likely all powered by switching regulators downstream of where you're measuring anyway.
The bigger problem in digital circuits is usually ground noise. So I'd keep focusing on ground issues even if it's just trying different star points, etc.
The inductors are unlikely to cause any damage. So try 'em if you want?
Probably not far off that!
its a Tek 5113 Dual Beam Storage scope. One of the old ones which stores waves on a CRT screen. Probably not bad when it was new but skip-napped it from Uni and its a little shakey lol.
I hope its not a problem with the scope, this would upset me as ive spent so much time nurturing it back to health
I will focus on shortening cables and rejig my grounding a bit, this will be a big task.
Ill try and find a way to put those inductors in too then if they cant do harm.
its a Tek 5113 Dual Beam Storage scope. One of the old ones which stores waves on a CRT screen. Probably not bad when it was new but skip-napped it from Uni and its a little shakey lol.
I hope its not a problem with the scope, this would upset me as ive spent so much time nurturing it back to health
I will focus on shortening cables and rejig my grounding a bit, this will be a big task.
Ill try and find a way to put those inductors in too then if they cant do harm.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.