Is it a DC brushless motor fan? (Most likely yes) They have their own DCDC switcher and generate back emf pulses.
Very interesting ... I switched the amp back on - with my CRO attached to the +ve DC rail of one of the SLBs.
Spiky ripple ... which went smooth as soon as I pulled that fan connector apart! 😱
So, yes, the fan is feeding crap back into the +ve DC rail.
So I will build your CLC filter onto the end of the DC supply to my fans.
For each: 2x 220uF/25v caps plus a 220uH inductor, with a 54.9 ohm res in parallel - sound OK?
Thanks,
Andy
Yes, give it a try. Maybe (optional) bypass the second cap with Wima MKS 2.2uF film cap.
Last edited:
Thanks, X.
Will do - but would there be any value (in terms of the end result) in increasing the value of the caps ... from 220uF, to say 680uF?
Andy
Sure, it will bring the corner frequency lower and farther away from the switch mode frequency so the attenuation is greater. I actually use 1mH ferrite core radial inductors because I happen to have them and 1000uF caps.
Aah, great, X. 🙂
I'll use 680uF caps but will stick with the 220uH inductors, due to space limitations in my cases. 🙁
Andy
Well, happy to report the CLC filter does remove the spikes that the fans were feeding back into each +ve DC rail. So the ripple on the +ve DC rails is now nice and smooth.
And this is without adding that Wima bypass cap you suggested (I forgot it, when I was building the filtered fan PS 😱 ).
Pic attached.
Andy
And this is without adding that Wima bypass cap you suggested (I forgot it, when I was building the filtered fan PS 😱 ).
Pic attached.
Andy
Attachments
Thanks, V. 🙂
What I find interesting is that the amps sound wonderful ... even with the spiky +ve DC rails! 😱
So, hopefully ... they'll sound even better now (when I've repeated the exercise for the 2nd amp).
Andy
Great work Andy! What are those beefy 120ohm resistors for? To slow the fans to make the fans even quieter?
Sort of, X. The +DC rails are ~21v; those 12w Mills reses reduce this to about 11v for the 12v fans. Then the auxiliary Noctua cable reduces this even further.
I should've probably just soldered the reses in place, with this rebuild of the fan power supply (instead of re-using the Phoenix connectors) ... but I guess I can do some experimentation with slightly larger-value reses, to see if I can get the fans quieter - without the mosfets getting too hot.
Andy
Last edited:
Incidentally, X - can you tell me what having the ~50 ohm res in parallel with the inductor does ... compared to not having it?
Thanks,
Andy
Thanks,
Andy
Just a quick question…
I can hear my toroid transformer hum when connected to the SLB and when adding the second SLB two secondary 3-4 the hum just gets louder.
My initial thought is I had possibly too much DC on the AC line but just a mere 1.5mV would never bother my toroidy transformer. I also checked some other things. If transformer is not connected on the secondaries, it is totally silent. Now I will try a different PSU for testing.
In the meantime, I’m thinking it is the snubber at the AC input on the PSU… I’ve added the default BoM values, but those are most likely not optimized at all for this transformer and it is making it hum. Do you think it is worth removing the snubber or I’m wasting time? I’m thinking wrong value here can only cause issues like ringing, etc…
Thanks
Do
I can hear my toroid transformer hum when connected to the SLB and when adding the second SLB two secondary 3-4 the hum just gets louder.
My initial thought is I had possibly too much DC on the AC line but just a mere 1.5mV would never bother my toroidy transformer. I also checked some other things. If transformer is not connected on the secondaries, it is totally silent. Now I will try a different PSU for testing.
In the meantime, I’m thinking it is the snubber at the AC input on the PSU… I’ve added the default BoM values, but those are most likely not optimized at all for this transformer and it is making it hum. Do you think it is worth removing the snubber or I’m wasting time? I’m thinking wrong value here can only cause issues like ringing, etc…
Thanks
Do
You can safely remove the snubber but I don’t think that’s the issue. Actually, the active bridge doesn’t need a snubber because the switching is timed to happen at zero crossover voltage vs 0.6v in a regular Si diode. That really pretty much eliminates the sudden switching of a diode that leads to ringing.
Is there any load when it hums/buzzes? It should not buzz without a load. Under heavy load (like 75% of rated VA, maybe some buzz is expected).
Usually buzzing of a trafo is DC in the toroidal or something wrong with the phase of polarity of the secondaries not being correct.
You cannot use the LT4320 with a center tapped transformer. Make sure your secondaries are not center tapped. They need to be independent windings.
Is there any load when it hums/buzzes? It should not buzz without a load. Under heavy load (like 75% of rated VA, maybe some buzz is expected).
Usually buzzing of a trafo is DC in the toroidal or something wrong with the phase of polarity of the secondaries not being correct.
You cannot use the LT4320 with a center tapped transformer. Make sure your secondaries are not center tapped. They need to be independent windings.
Yep! Transformer is quad secondary, no center tap.
I found it weird because these Toroidy transformer have always been super quiet. As for DC, I’m reading no more than 2mV, which is quite low.
Yes, transformer does not hum with no load (with or without PSU).
I’ll keep searching
Thanks
Do
I found it weird because these Toroidy transformer have always been super quiet. As for DC, I’m reading no more than 2mV, which is quite low.
Yes, transformer does not hum with no load (with or without PSU).
I’ll keep searching
Thanks
Do
Hi X,
I had already tried flipping one of the phases but no difference. I’ve just removed the snubber and also nothing changed.
I will try to install a DC blocker for testing and some other things. Then I will contact Toroidy about this. I never had any issues with them in the past so this seems not normal.
Thanks
Do
I had already tried flipping one of the phases but no difference. I’ve just removed the snubber and also nothing changed.
I will try to install a DC blocker for testing and some other things. Then I will contact Toroidy about this. I never had any issues with them in the past so this seems not normal.
Thanks
Do