| ElectroAcoustic |
Hello,
I just realized how rusty I am on power supplies, so I thought I would ask a few questions here.
Say I have a simple power supply for an amplifier. Fused 120v 60Hz input, a transformer, a bridge rectifier, and two filter caps. Values not really important, but I have a few questions:
1. Could I power this up with no load attached, without harming anything?
2. Would the filter caps charge?
3. Would the filter caps DIScharge, when power was disconnected?
4. How do the filter caps discharge, i.e. what is the path that the discharge would follow? Through the grounded side of the caps?
Thanks in advance for any input. This is perplexing me, and any insight would be greatly appreciated.
Thanks,
ElectroAcoustic |
|
|
| AndrewT |
Hi,
1. Yes, no harm will ensue.
2. Yes, open circuit voltage is higher than loaded voltage.
3. Yes, but very slowly. Could take days/weeks.
4. Leakage through the diodes and through the capacitiors. |
|
|
| jaap_vk |
1. Yes. Only beware that rectified volatge could be up to two times that of the voltage of your transformer. Make sure the caps can handle that. If not .......boom
2. Yes. See above
3. Only very slowly. Thiswould be leaking current of the caps. With good quality caps this could take hours if not days.
4. To make sure they discharge just add a resistor parallel to the caps |
|
|
| ElectroAcoustic |
Hello,
Thank you for the replies.
So when designing a power supply into an amplifier, is it necessary to design a seperate circuit to drain the caps when power is removed?
In power supply schematics, there are no resistors parallel to ground between the bridge and the load, on either side of the caps.
Is there just a resistor to ground? Sometimes I see a Zener in series with a resistor, in an amp schematic between the rails and ground. |
|
|
| ElectroAcoustic |
Well, another reason I ask, is that a project I am working on, the neg. rail is not draining properly. It is erratic, and is causing the input op-amp to bleed noise into the speakers when powered down.
It's driving me crazy.
When powered up, all is perfectly fine and stable. Just on shut down, I get a horrible screeching, and I know it's because the neg rail isn't draining properly.
Any ideas? |
|
|
| BWRX |
| quote: | Originally posted by ElectroAcoustic
So when designing a power supply into an amplifier, is it necessary to design a seperate circuit to drain the caps when power is removed? |
It's not necessary, but it's good for safety reasons (especially if you're dealing with high voltages) and to increase the life of the caps.
| quote: | Originally posted by ElectroAcoustic
Well, another reason I ask, is that a project I am working on, the neg. rail is not draining properly. It is erratic, and is causing the input op-amp to bleed noise into the speakers when powered down. |
This is because your supply rails are discharging at different rates. That doesn't mean the negative rail is not draining properly. To avoid the noise from the op amp getting through to the speaker you can do a number of things. The simplest (I guess it would really depend on the circuit though) is to make sure the power for the output stage dies before the power to the input stage. You could also put a relay between the output stage and the speaker and open the relay before the rails discharge. Or you could come up with some kind of muting circuit that disables the output stage before the rails discharge. |
|
|
| AndrewT |
Hi,
I wonder if an input mute would kill the sckreech when powering down?
Easy to check with a shorting plug in each RCA. |
|
|
| ElectroAcoustic |
| So what could cause the rails to discharge unevenly? Mismatched components? |
|
|
| AndrewT |
unequal draw designed into the amplifier.
standard range of tolerance on components. |
|
|
| ElectroAcoustic |
I forgot to mention that the input is a balanced line, xlr type input. The first stage, and the one causing the source of the noise, is the differential input op amp. Immediately after the diff amp, the signal passes through an input attenuator pot, where it then routes to the respective amp channel. There is no muting circuit in this amp.
The problem did not used to be there, so I don't believe that the uneven rail collapse is caused by a design fault. Am I correct to assume, then, that some component may be out of tolerance?
I replaced the filter cap on the neg rail, along with the bridge diodes, and the problem persists.
To go out on a limb, is there a relay, say a DPST NO type, that can be inserted in the 120v AC line after the master power switch, that could lift the +-Vs lines into the input differential op amp when the amp is off, or maybe lift the output of the diff amp?
Not sure if it would be that simple. |
|
|
| AndrewT |
| quote: | Originally posted by AndrewT
unequal draw designed into the amplifier.
standard range of tolerance on components. |
| quote: | Originally posted by ElectroAcoustic
.........problem did not used to be there, so I don't believe that the uneven rail collapse is caused by a design fault. Am I correct to assume, then, that some component may be out of tolerance?
I replaced the filter cap on the neg rail, along with the bridge diodes, and the problem persists.
To go out on a limb, is there a relay, say a DPST NO type, that can be inserted in the 120v AC line after the master power switch, that could lift the +-Vs lines into the input differential op amp when the amp is off, or maybe lift the output of the diff amp? |
Not a design fault, but designed to be different. It is very difficult to get an equal draw of current from both rails of a ClassA voltage amplifier.
The standard tolerance of electrolytics is enormous. they come that way as standard.
If you cannot see how to wire up a mains relay then I suggest you get trained/qualified person to do the mains alteration for you. |
|
|
| cliff |
You could also try putting a bleed resistor across the -ve rail cap to speed up its discharge.
try a 1K or so (you'll have to calculate the power dissipation -we are not going to do all the work for you :) ) and see if it makes any difference. |
|
|
| DJ Exprice |
| quote: | Originally posted by jaap_vk
4. To make sure they discharge just add a resistor parallel to the caps |
:eek: WAIT! Before you do that, look at this diagram that I made:
Use it with the caps!
(Use AC for the I/O input - from right before the transformer (120 users only) or after the transformer (AC ONLY!))
If you use the DC for the I/O, the discharge will take a long time. You could use a SMPS AC or DC if when you switched the amp off the voltage from it hits 0v. :smash: Not recommended, however. AC relays are easier, and cheaper...and safer than SMPS circuits :) |
|
|
|
