Strange ripple on rails when amp is connected

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

My speakers are humming when connected to my last diyamp, the frequency of the hum is approx 160Hz, measured at the speaker output.

After a lot of testing and measuring, I can say the hum is comming from the power supply. When connecting my scope between ground and the + or - rail, I'm measuring a ripple with a frequency of +-320Hz and 60mVpp, see the picture.

When disconnecting the amp the ripple on the rails is gone. I tried an other transformer, rectifier bridge and caps, nothing changed this phenomenon.

The setup of the power supply is:
1. mains 240Vac/50Hz
2. EMI filter
3. soft start
4. Amplimo transformer 625VA/ 2x35V
5. rectifier bridge made with 4x HFA25PB60
6. 10000µF+4700µf +0.11Ohm/10W + 10000µF
7. amplifier


Can anybody tell me what could causes this ripple and what I could do to remove it?

Thanks
 

Attachments

  • 320Hz - 60mVPP.JPG
    320Hz - 60mVPP.JPG
    179.4 KB · Views: 138
  • SAM_0790.JPG
    SAM_0790.JPG
    250.2 KB · Views: 130
Mooly and Tauro0221,
Funny, I din't think on that before. I connected the scope to my frequency generator, and indeed, the frequency reading of on the scope was totaly wrong. After cleaning the contact of the measuring cables/interfaces scope, all is fine.

Checked again the ripple ==> when input of amp is shorted

- ripple on rails = 60mVpp /100Hz
- speaker output = 10mVpp / 50Hz



DF96,

Poor PSSR, I don't think that it will differ much with many other complementary symmetric amps. Or am I missing something?? Please have a look at the shematic below. I already changed the Fet drivers to BJT's, but his hasn't made a difference.
 

Attachments

  • Final schematic.pdf
    28.9 KB · Views: 59
Administrator
Joined 2007
Paid Member
I too think layout and grounding is probably the most likely cause and from what I can see from your picture it all looks really neatly built.

If the "hum" has a harsh sounding edge to it (harmonics present) then grounding is a prime suspect.

50 hz hum. If it were induced hum then I would think the output would be more sinusoidal and tbh at that level probably not audible. 50 hz as a sinewave is a very deep pure tone (if the level is high enough).

I really don't know what to suggest without first hand knowledge of all the wiring layouts. Every conductor has to be treated as a resistance and you have to ask that "if a current flows and volt drop develops", can that get into the signal path.

Probably not much help but this might give you ideas. You'll have to read it all as the thread wandered a bit.

http://www.diyaudio.com/forums/solid-state/101321-3-stage-lin-topology-nfb-tappings.html
 
I too think layout and grounding is probably the most likely cause and from what I can see from your picture it all looks really neatly built.

If the "hum" has a harsh sounding edge to it (harmonics present) then grounding is a prime suspect.

I agree, see also this thread. On symmetrical PSUs with single bridge and centre tap as GND, where that centre tap connects with respect to signal reference (GND) is all important.

50 hz hum. If it were induced hum then I would think the output would be more sinusoidal and tbh at that level probably not audible. 50 hz as a sinewave is a very deep pure tone (if the level is high enough).

Worth an experiment.
I have a DIY class A headphone amp that is powered by a toroid. Because of limited space, it was near the amplifier board and with no signal on the input I could hear a very faint hum. After hanging the toroid on the outside of the case, the hum disappeared.

Move all transformers away from the wiring through which the signals flow as far as possible. If hum goes away, it was induced.
 
Problem solved

1: after the right frequency measurements of 50Hz and 100Hz, I stopped thinking that the problem was the power supply

2: as mentioned by Nigelwright7557, I put in a 33R before the cap that buffers the LTP's and VAS ==> humm was still there, but amp seems to be having more detail

3: I splitted the audio ground and power ground by a 10R resistor.
==> this solved the humming problem.

With the input shorted and my ear against the tweeter, I can't hear a thing!

Thanks guys!
 
3: I splitted the audio ground and power ground by a 10R resistor.
==> this solved the humming problem.

With the input shorted and my ear against the tweeter, I can't hear a thing!

Thanks guys!

Classic ground loop problem.

I built a mixer on a pcb with no thought to grounds and it hummed really badly. When I split off the power supply and had one star ground the humm went completely.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.