The caps across the rectifiers might increase noise, but they won't affect hum.
You have two points marked common ground. How are these connected? The two channels of a stereo system normally have their grounds connected together somewhere, and this can introduce a loop.
What is your measured and calculated supply ripple, circuit PSRR and output hum? Simply saying you have hum does not tell us very much. We need numbers.
If the problem is supply ripple, then adding capacitance might help. If the problem is induction into a loop, then adding capacitance could make things worse.
As Andrew has asked, is this now exactly what you have built?
BTW you have used more bridges than necessary.
You have two points marked common ground. How are these connected? The two channels of a stereo system normally have their grounds connected together somewhere, and this can introduce a loop.
What is your measured and calculated supply ripple, circuit PSRR and output hum? Simply saying you have hum does not tell us very much. We need numbers.
If the problem is supply ripple, then adding capacitance might help. If the problem is induction into a loop, then adding capacitance could make things worse.
As Andrew has asked, is this now exactly what you have built?
BTW you have used more bridges than necessary.
All the above are mounted on a big PCB. So no wires. When you say that the capacitors will increase the noise what do you mean? Also to me the problem is a noise listenable only from the woofer and not from the midrange. It is clearly a 50 - 100 Hz. Maybe for the amperes I need more capacitance than 130000uf. I can not measure ripple sorry
thanx
thanx
In the past people made perfectly good systems with 5000-10000uF caps, so it is very unlikely that you have too little capacitance. By my calculations, assuming 4A load, your PSU should provide about 15mV ripple - very low!
What test equipment do you have? You can't debug a circuit just using your ears.
What test equipment do you have? You can't debug a circuit just using your ears.
I have found this. Is that what you have built? Note that fig 3 is wrong - see comment underneath about C4. JLH says 33mF is fine for a simple PSU, although with some hum. He recommends a regulated PSU.
The negative end of C4 is the reference point for the NFB, so it must go to a clean reference point such as the signal ground of whatever comes before the amp.
The negative end of C4 is the reference point for the NFB, so it must go to a clean reference point such as the signal ground of whatever comes before the amp.
JLH says about 3mV hum from his simple supply. Your supply shouldl have much less hum than his, so PSU ripple is probably not the cause of your hum. Maybe time to start a new thread about a hum problem, not a PSU problem. However, with not much test equipment you might not get very far. Can you fix your scope - much easier then.
Another thing I noticed is that when in romove the resistance from RC filter in the positive rail (only caps) nothing different is happening, but when I do this in the negative rail then an excessive ripple noise is heard. TGhis is the case for both channels. Why this difference between rails is happening I do not know. Any ideas?
As you can spot I have a commoon ground that is the connection of the regulated and non regulated supplies. The feedback capacitor defenitely goes to this common ground. The unregulated -V goes in to the emmiter resistor of the output transistor. What can I do to make better PSRR? Thanx for the help DF96!
Another fact is that I measured yesterday the voltage between the chassis and the signal ground to be about 14mV ac. I would like to mention here that the chassis is connected to power ground via a 10 ohm resistor and a cap in parallel. Is that normal? Is maybe the cause of the hum. Also as it is obvious the chassis is connected to the outlet ground of the house.
14mV across 10 ohms is 1.4mA which seems rather high, unless you have several items of modern equipment connected each contributing some leakage current from their mains filter capacitors. However, this may not be the source of the hum as everything ought to be referred to the signal ground.
Having confirmed that your circuit has poor PSRR on the negative rail, you really need advice from someone who has successfully built the JLH class A circuit. You probably have a grounding problem.
Having confirmed that your circuit has poor PSRR on the negative rail, you really need advice from someone who has successfully built the JLH class A circuit. You probably have a grounding problem.
14mVac between Signal Ground and Chassis indicates an earth leakage current of ~ 0.014/10 ~ 1.4mAac
This is not excessive but more than normal. Usually one cannot measure (<0.1mVac narrow band on a DMM) the AC voltage across the Disconnecting Network between Signal Ground and Power Ground.
However, have you connected Power Ground to Signal Ground.
The leakage current from Power Ground to Chassis could be contaminating the Signal Ground connection route. The amplifier then multiplies this up to appear as hum at the speakers.
This is not excessive but more than normal. Usually one cannot measure (<0.1mVac narrow band on a DMM) the AC voltage across the Disconnecting Network between Signal Ground and Power Ground.
However, have you connected Power Ground to Signal Ground.
The leakage current from Power Ground to Chassis could be contaminating the Signal Ground connection route. The amplifier then multiplies this up to appear as hum at the speakers.
Identify the signal circuit route. All the way from the Source Hot pin back to the Source Cold pin. Find this route in the actual amplifier. Draw it out as it is actually assembled in the amplifier. Mark this route in Red
Now identify the Power circuit route. Go through the same steps as for the Signal route. Mark this route in Black
Finally identify the speaker circuit route do the same again and mark this route in Green.
Are there any traces or wires that share their length, or just part of their length between any two of these circuits? If there is, you must separate them, particularly the Signal Route.
Now identify the Power circuit route. Go through the same steps as for the Signal route. Mark this route in Black
Finally identify the speaker circuit route do the same again and mark this route in Green.
Are there any traces or wires that share their length, or just part of their length between any two of these circuits? If there is, you must separate them, particularly the Signal Route.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.