What a drag that your transistors were fake!! Care to name the crooked supplier? Have you considered using MJL4281/4302 transistors as replacements? They're quite robust; I got mine directly from ON Semiconductor, so there's probably little chance of getting fakes.
Here's a TIP41C mounted:
Heatsink-paste-mica-paste-transistor-bush-washer-nut
(I will be cleaning up the excess paste...)
Heatsink-paste-mica-paste-transistor-bush-washer-nut
An externally hosted image should be here but it was not working when we last tested it.
(I will be cleaning up the excess paste...)
Not fun to spend good money and receive useless *insert word here*.
Unfortunately I have no idea who the supplier is. I bought from a very well-known distributor in South Africa: Communica. I've received good stuff from them in the past.
There isn't much choice in South Africa, unless you are willing to lay out a lot of cash, or willing to risk the fakes. I was going to use this amplifier in bridge mode, but I've taken the direction of bi-amplification, so the smaller transistors will do fine (and I've used them plenty and love them). Else I would have gone with TIP35/6.
For the ones you mention, they're R110 a pop (about $7), which works out to almost R900 ($55), just for the 4 pairs, and that's massive money for me. The TIPs cost me R80 ($5) for the lot. And those you mention are completely overkill for this size amplifier. 🙂
"those you mention (MJL4281/4302 transistors) are completely overkill for this size amplifier."
Yeah....,.that's what audiophiles (and musicians) do....OVERKILL!!!
Yeah....,.that's what audiophiles (and musicians) do....OVERKILL!!!
Sigh. This amplifier hums. I'm going to have to take a scope to it. I'm not sure where the hum is coming from. Measured hum on the output is 10 mV (rms). Changing the resistor in the LTP to a current source makes absolutely no difference whatsoever. I've changed the grounding, tried to ground the output on the amplifier, not the star, moved the earth connection, moved the transformer center-tap, decreased the impedance between the common connections of the PS capacitors. The hum is constant, and doesn't change at all.
Still, my one light at the end of a seemingly infinite tunnel is that the hum is much less on the HF channel than the LF channel, meaning it's likely to be all in the preamplifier. I'm going to disconnect the output of the preamp from the input of the amplifiers and see what happens.
I've drawn some wiring diagrams below (mostly so I can ponder my wiring). Any thoughts would be most welcome! 🙂
Still, my one light at the end of a seemingly infinite tunnel is that the hum is much less on the HF channel than the LF channel, meaning it's likely to be all in the preamplifier. I'm going to disconnect the output of the preamp from the input of the amplifiers and see what happens.
I've drawn some wiring diagrams below (mostly so I can ponder my wiring). Any thoughts would be most welcome! 🙂
An externally hosted image should be here but it was not working when we last tested it.
Have you shorted the input sockets with zero ohm dummy plugs yet?
post a detailed pic of the PSU.
post your sch that matches that PSU.
post a detailed pic of the PSU.
post your sch that matches that PSU.
Yes, changing the input doesn't make a difference, short, signal, open.
Physical PSU (more or less - diodes to preamplifier PSU are omitted - they're there in real life):
An externally hosted image should be here but it was not working when we last tested it.
Schematic:
An externally hosted image should be here but it was not working when we last tested it.
Oh, and I've connected the star directly to where the two capacitors from which I take the DC connect to each other with a very thick cable.
And it may be worth mentioning - the hum is 100Hz, so my AC is connected correctly.
Last edited:
leave the zero ohms dummy load in place until you have sorted the power amplifier wiring.
The Main Audio Ground MUST NOT be taken to the PSU capacitor link.
Let us see a pic.
The Main Audio Ground MUST NOT be taken to the PSU capacitor link.
Let us see a pic.
Thanks for the interest. Here's a photo, but it's slightly outdated. The labels tell everything. All DC power going to amplifiers and the preamplifier are about 13 AWG solid copper.
An externally hosted image should be here but it was not working when we last tested it.
That makes correcting the wiring error quite difficult.
I would disconnect the whole PSU from the amplifier and try a completely independent PSU and see if that measures no hum.
I would disconnect the whole PSU from the amplifier and try a completely independent PSU and see if that measures no hum.
It seems odd to me that you have gone to GREAT length to filter the power supply (16 x 3,300uf!!!) but then do not regulate it at all for the power amp load and provide only noisy Zener diode regulation for the preamp. I think it would be better to regulate the power amp supply with a LT1083 and then LM317/337 for the preamp. I don't know if this would solve the hum problem, though.
Although the power supply is on board, it is only connected to anything via 1 ground connection. All other DC is taken with wires.
I went to great length to filter the power because this was going to be a bridge configuration, and the ripple at maximum load needed to be minimized as much as possible because in bridge configuration, the effect of ripple when approaching maximum output is doubled, which causes too much clipping. I only need the zeners because i needed a reliable supply voltage, not a necessarily clean one. I was originally going to use a separate transformer.
The amplifier and preamplifier have theoretically adequate power supply rejection, and at about 3.5 A load, ripple should be less than 0.5 Vp-p.
I have some ideas, but i can't try them now. But i am basically going to disconnect the pcb part of the supply.
I went to great length to filter the power because this was going to be a bridge configuration, and the ripple at maximum load needed to be minimized as much as possible because in bridge configuration, the effect of ripple when approaching maximum output is doubled, which causes too much clipping. I only need the zeners because i needed a reliable supply voltage, not a necessarily clean one. I was originally going to use a separate transformer.
The amplifier and preamplifier have theoretically adequate power supply rejection, and at about 3.5 A load, ripple should be less than 0.5 Vp-p.
I have some ideas, but i can't try them now. But i am basically going to disconnect the pcb part of the supply.
Hi,
Just as a test. I am going to repeat it again just as a test, remove the earth ground and see if the hum goes away.
Just as a test. I am going to repeat it again just as a test, remove the earth ground and see if the hum goes away.
"I was originally going to use a separate transformer."
That would be a VERY good idea, although a transformer with separate secondary windings for the power and pre sections would work just as well, if there is enough VA for each.
That would be a VERY good idea, although a transformer with separate secondary windings for the power and pre sections would work just as well, if there is enough VA for each.
On most of my amps crossover distortion goes at around 10mA.
I believe Peavey were also fans of low bias currents.
Why waste power for no benefit ?
Thanks. I already tried it, no difference.
What I am going to do is remove the ground connection from the star to the capacitors. I have a connection from closer to where the DC comes off the capacitors.
With this amplifier I've tried a different (and more conventional) grounding to what I've done in the past. I just haven't had time to do everything I want to try to remove the hum yet.
I usually set my amps up to about 20 mA, or where turning the bias just starts to make massive jump in bias current, and then back ever so slightly. This has made crossover distortion vanish, and all these amplifiers have been stable (thermally). And I agree with this method. But with this one, I have to simply turn it on, and it's ready to go. No setup required, and to me that's a big bonus because I have 4 amplifiers. And the wasted power isn't all that wasted - it's only 15 W per amplifier, and I get about 1.5 W class A (about 94 dB SPL on my speakers).
ETA: Yes! And there is another thing I have heard which I love - with the larger bias, turn on and off thumps are much weaker than my amplifiers with less bias. Also, when you turn it off, it makes a turn off thump as you flip the switch, so the thump doesn't happen a few seconds after turn off. This is a very nice characteristic to me.
Last edited:
did you read post11?
If you require an average of 94dB output, then for a peak to average ratio in your music of only 10dB you would need 104dB while still in ClassA, that would require 15W of ClassA to allow you to listen @ 94dB and remain in ClassA most of your listening time.
Or looked at from the other direction. If you have peaks that just reach 0.7W and stay in ClassA, then your average listening level would be between 70mW (10dB peak to average) and 7mW (20dB peak to average).
Last edited:
Yes, and it's fine with me. I have got a few ideas about that bias, but I'm not worried about it. First, the hum.
Ye, ye. 😉 Still, the lack of setup is worth a lot to me.
- Status
- Not open for further replies.