Hi,
post38calls up diagram 3.
That is a dual rectifier that is used with a dual secondary. The transformer is NOT connected to ground. But your post36 diagram shows the transformer centre tap connected to that offending plate.
One cannot mix these two different PSU arrangements.
If you adopt different supply rails then I suggest you ONLY use a dual rectifier version and keep the multi-channel audio grounds separate from each other.
I have found it best to use a pair of secondaries dedicated to each channel. But these 4 secondary transfomers (for stereo) are not generally available cheaply.
Note also that post36 shows the RCA ground going to the PCB. and the PCB ground going to the audio ground. Where is the signal ground? It can return direct from RCA ground to audio ground but NOT if the PCB has a common clean and dirty ground. Unless you can cut the clean to dirty track on the PCB.
post38calls up diagram 3.
That is a dual rectifier that is used with a dual secondary. The transformer is NOT connected to ground. But your post36 diagram shows the transformer centre tap connected to that offending plate.
One cannot mix these two different PSU arrangements.
If you adopt different supply rails then I suggest you ONLY use a dual rectifier version and keep the multi-channel audio grounds separate from each other.
I have found it best to use a pair of secondaries dedicated to each channel. But these 4 secondary transfomers (for stereo) are not generally available cheaply.
Note also that post36 shows the RCA ground going to the PCB. and the PCB ground going to the audio ground. Where is the signal ground? It can return direct from RCA ground to audio ground but NOT if the PCB has a common clean and dirty ground. Unless you can cut the clean to dirty track on the PCB.
blap0220 said:
I'm not familiar with the details of this design, so can you tell me are the input ground, and output ground connected on the circuit board?
Are your input jacks isolated from chassis ground?
Note that this design has no resistance or decoupling between the power/driver stages and the sensitive input stages so it will be more sensitive to B+/B- lead inductance. Also there is no HF power supply decoupling on the board, unless I missed it. You might try .1uFs from B+/B- to the output ground, not input ground.
Might also try .1uFs across the main PSU caps.
Looking at this again, I'm fairly certain that it is input to output stage coupling at HF.
Pete B.
Weird Problem!
OK!!!! I got a new problem. OR this might be the cause of my previous problem. And it's weird!! At least to me it is.
So I'm now using one positive supply rail and one negative supply rail to supply both amps. Whenever I lift one of the negative DC supply wires, I hear "HUMMMMMM". It's quite loud. But when I move it back down, the hum goes away. I thought it might be because of a certain component the wire comes close to, but it doesn't seem like it. Because the same thing happens when I lift the other negative DC wire (two negative DC wires from same cap bank). It seems more like once the wires reach a certain height or position, the hum starts.
And it's very discrete. I move it 1mm up past the hum line and it starts, but then I move it back down and it goes away. WEIRD!!! to me. Is this oscillation?! Is it my filter caps?
I have a feeling this is where my problem was all along.
THANKS!
OK!!!! I got a new problem. OR this might be the cause of my previous problem. And it's weird!! At least to me it is.
So I'm now using one positive supply rail and one negative supply rail to supply both amps. Whenever I lift one of the negative DC supply wires, I hear "HUMMMMMM". It's quite loud. But when I move it back down, the hum goes away. I thought it might be because of a certain component the wire comes close to, but it doesn't seem like it. Because the same thing happens when I lift the other negative DC wire (two negative DC wires from same cap bank). It seems more like once the wires reach a certain height or position, the hum starts.
And it's very discrete. I move it 1mm up past the hum line and it starts, but then I move it back down and it goes away. WEIRD!!! to me. Is this oscillation?! Is it my filter caps?
I have a feeling this is where my problem was all along.
THANKS!
Re: Weird Problem!
It would help if you answer the previous questions, but yes, the problem in likely in the V+/V- supply lines, which is influenced by the grounds. You have a high freq oscillation due to poor supply decoupling in that design making it sensitive to lead inductance. Not good that your amp is oscillating, could damage speakers, amp, etc.
Is this your amp, Figure 1B? http://sound.westhost.com/project03.htm
Pete B.
blap0220 said:
It would help if you answer the previous questions, but yes, the problem in likely in the V+/V- supply lines, which is influenced by the grounds. You have a high freq oscillation due to poor supply decoupling in that design making it sensitive to lead inductance. Not good that your amp is oscillating, could damage speakers, amp, etc.
Is this your amp, Figure 1B? http://sound.westhost.com/project03.htm
Pete B.
Re: Re: Weird Problem!
It's the updated version of that project. The P3A.
http://sound.westhost.com/project3a.htm
In the supply decoupling? Could it be due to bad capacitors?
This is my power supply design. Diagram 3.
http://www.tnt-audio.com/clinica/ssps1_e.html
Thank you
PB2 said:
It's the updated version of that project. The P3A.
http://sound.westhost.com/project3a.htm
In the supply decoupling? Could it be due to bad capacitors?
This is my power supply design. Diagram 3.
http://www.tnt-audio.com/clinica/ssps1_e.html
Thank you
Re: Re: Re: Weird Problem!
Did you build it from the kit or your own parts?
I would check CV+ and CV- on both boards, that they are
the correct value, and all other caps.
Can you determine if the input and output grounds are connected on the circuit board?
Do your input jacks have isolated grounds?
So the supply is not one for each side, rather one bridge for pos and one for neg. This is fine, have you checked the voltages with a meter? Actually, looking at your grounding/PSU diagram it looks as if you have one bridge and caps for each channel, making it different from diagram 3?
There is no output inductor in that amp - what speaker cable are you using, and you might try different speakers. The amp probably does not do well with highly capacitive speaker loads.
Pete B.
blap0220 said:
Did you build it from the kit or your own parts?
I would check CV+ and CV- on both boards, that they are
the correct value, and all other caps.
Can you determine if the input and output grounds are connected on the circuit board?
Do your input jacks have isolated grounds?
So the supply is not one for each side, rather one bridge for pos and one for neg. This is fine, have you checked the voltages with a meter? Actually, looking at your grounding/PSU diagram it looks as if you have one bridge and caps for each channel, making it different from diagram 3?
There is no output inductor in that amp - what speaker cable are you using, and you might try different speakers. The amp probably does not do well with highly capacitive speaker loads.
Pete B.
Re: Re: Re: Re: Weird Problem!
I used his PCB to build the amps.
I am using the diagram 3 PSU. One rectifier for pos and one for neg. I'm not using what I had drawn previously.
If it's related to the Capacitors on the amps, then should I be seeing this inductance "hum" when I mess with the positive DC leads as well?
Yes, the input grounds are isolated. I am using an acrylic back panel. Right now, I have the input grounds bypassing the amp grounds and going directly to my star ground.
PB2 said:
I used his PCB to build the amps.
I am using the diagram 3 PSU. One rectifier for pos and one for neg. I'm not using what I had drawn previously.
If it's related to the Capacitors on the amps, then should I be seeing this inductance "hum" when I mess with the positive DC leads as well?
Yes, the input grounds are isolated. I am using an acrylic back panel. Right now, I have the input grounds bypassing the amp grounds and going directly to my star ground.
Re: Re: Re: Re: Re: Weird Problem!
The negative lead supplies the diff pair tail, the VAS load, and the negative output stage. Note C6 on the negative side of the output stage, NPN and PNP devices are not exact compliments and he probably added that cap to control an oscillation in the negative side of the output stage. Check C6 is it really 100 pF?
See: http://xtronics.com/kits/ccode.htm
Might check for a miswire in the PSU or any wrong value component. Just because it powers up does not mean that the PSU is wired correctly and can supply current under load.
Hard to help further without trying it say with a resistive dummy load on the output, using a scope, and without knowing the board layout and grounding paths.
Pete B.
blap0220 said:
The negative lead supplies the diff pair tail, the VAS load, and the negative output stage. Note C6 on the negative side of the output stage, NPN and PNP devices are not exact compliments and he probably added that cap to control an oscillation in the negative side of the output stage. Check C6 is it really 100 pF?
See: http://xtronics.com/kits/ccode.htm
Might check for a miswire in the PSU or any wrong value component. Just because it powers up does not mean that the PSU is wired correctly and can supply current under load.
Hard to help further without trying it say with a resistive dummy load on the output, using a scope, and without knowing the board layout and grounding paths.
Pete B.
My ground paths are pretty much still the same as what I had drawn in my picture. But I've moved the connections on the ground plate around to see if they made a diff. Doesn't the current flow torward/into earth (chassis)? So if that's the case, the cleanest signals should be furthest away from the wire connecting that ground plate to the chassis right?
I'm pretty sure the C6 capacitor value is correct (Mica).
Also, if the oscillation is only in one channel, will I hear it in the other channel as well? I hear it in both channels. One is maybe stronger than the other
I'm pretty sure the C6 capacitor value is correct (Mica).
Also, if the oscillation is only in one channel, will I hear it in the other channel as well? I hear it in both channels. One is maybe stronger than the other
blap0220 said:
Your grounding looks basically OK, based on what I see there, but there are more details involved.
It's hard to say what to do at this point without modding the amp for better immunity to stray effects. No the current should not flow to the earth unless their is a fault condition. Think of the enclosure as simply a shield. Think of where high current flows, PSU caps, output transistors, output lead wire, speaker, output ground wire, transformer center tap/capacitor ground. This is the main high current path. It is hard to say what to do since many factors come into play at HF, the type of capacitor that you used for CV+ and CV- for example. A stacked type has the lowest ESL, and ESR. What type did you use?
It's really difficult to help more without more details, CFP output stages are fussy, and the design does not seem to have many precautions to combat the tendency to oscillate. Then again we also don't know the details of your construction.
Pete B.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.