I've built the single ended mosfet amp design following the build here:
DIY Class-A 2SK1058 MOSFET Amplifier Project (but using a 24v SMPS for the power side).
It's my first foray into solid state, and I must say the sound output is VERY pleasant (relative to my various PP and SE vacuum tube builds AND my Marsh A200 - not better, not worse, just dang good).
But I have 3 issues I'd like to work on:
1. I get 10 - 11 vdc on the outputs, and don't understand why given the caps on the outputs (which are installed correctly) and no obvious other sources of voltage. The outputs are isolated from the chassis and (the negatives) are tied to a star ground. Inputs are similarly well isolated. What could cause this?
2. There's a turn-on and turn-off thump which I'd like to get rid of. Again, not sure why given the PS de-coupling. I could increase the B+ de-coupling caps (at the 24V point on the signal end of the schematic) but not sure those caps will get rid of the thumps.
3. I'd like to modify this to use following my tube linestage (which has a volume/attenuation control). It looks to me like this amp has a 50k input impedance WITH the volume control. How do I modify this to remove the volume control yet maintain a reasonable input impedance for the amp using my particular preamp (no trafos on the preamp outputs and using 3S4 tubes). Replacing with a 50k resistor would seem to attenuate all the signal per the design volume control leaving me with no signal and virtually nothing out of the speakers.).
I'm really only beginning to understand input and output impedance, so please know that it may take me a while to understand any replies.
Thanks all,
Carl
DIY Class-A 2SK1058 MOSFET Amplifier Project (but using a 24v SMPS for the power side).
It's my first foray into solid state, and I must say the sound output is VERY pleasant (relative to my various PP and SE vacuum tube builds AND my Marsh A200 - not better, not worse, just dang good).
But I have 3 issues I'd like to work on:
1. I get 10 - 11 vdc on the outputs, and don't understand why given the caps on the outputs (which are installed correctly) and no obvious other sources of voltage. The outputs are isolated from the chassis and (the negatives) are tied to a star ground. Inputs are similarly well isolated. What could cause this?
2. There's a turn-on and turn-off thump which I'd like to get rid of. Again, not sure why given the PS de-coupling. I could increase the B+ de-coupling caps (at the 24V point on the signal end of the schematic) but not sure those caps will get rid of the thumps.
3. I'd like to modify this to use following my tube linestage (which has a volume/attenuation control). It looks to me like this amp has a 50k input impedance WITH the volume control. How do I modify this to remove the volume control yet maintain a reasonable input impedance for the amp using my particular preamp (no trafos on the preamp outputs and using 3S4 tubes). Replacing with a 50k resistor would seem to attenuate all the signal per the design volume control leaving me with no signal and virtually nothing out of the speakers.).
I'm really only beginning to understand input and output impedance, so please know that it may take me a while to understand any replies.
Thanks all,
Carl
Hi.
1) Is this measured with a load, open circuit with a modern meter i would wxpect to see voltage, try putting a 300ohmish resistor from the output to ground.
2) This circuit will have a turn on/off thump by its nature, the only way around is controlling the rise and fall time of the PSU which wont be easy.
3) conect a 50k resistor across the input connector and then connect the wiper ind of the 0.47uF cap to the input.
HTH
Chris
1) Is this measured with a load, open circuit with a modern meter i would wxpect to see voltage, try putting a 300ohmish resistor from the output to ground.
2) This circuit will have a turn on/off thump by its nature, the only way around is controlling the rise and fall time of the PSU which wont be easy.
3) conect a 50k resistor across the input connector and then connect the wiper ind of the 0.47uF cap to the input.
HTH
Chris
1. If you are measuring with no load other than your meter you will see DC since you are charging the cap through a high resistance. I assume that your speaker is still alive, do you see high DC after the amp has been on a few seconds? (use an 8-10R resistor for testing)
2. Try removing the 33 uf cap. It creates a turn on delay for the mosfet with the resistance of the bias pot. The bias voltage on the mosfet rises slowly, leaving the Drain at rail voltage until it turns on. Without the 33 uf cap the output cap will be charged by a lower voltage which may reduce the thump. Otherwise a relay that shorts the output momentarily will allow the output caps to charge more quickly.
3. Remove the pot and connect the input to the 0.47 uf cap. Input impedance will be a megohm plus the part of the bias pot from wiper to ground. Use a 100-220K resistor across the input to make the impedance a bit more practical. See the Zen input arrangement. http://passdiy.com/pdf/zenamp.pdf
2. Try removing the 33 uf cap. It creates a turn on delay for the mosfet with the resistance of the bias pot. The bias voltage on the mosfet rises slowly, leaving the Drain at rail voltage until it turns on. Without the 33 uf cap the output cap will be charged by a lower voltage which may reduce the thump. Otherwise a relay that shorts the output momentarily will allow the output caps to charge more quickly.
3. Remove the pot and connect the input to the 0.47 uf cap. Input impedance will be a megohm plus the part of the bias pot from wiper to ground. Use a 100-220K resistor across the input to make the impedance a bit more practical. See the Zen input arrangement. http://passdiy.com/pdf/zenamp.pdf
Thanks a ton Chris and Bob! These are very helpful answers.
The voltage on the outputs was, indeed, measured without a load using a Fluke 12B meter (which I wouldn't do with tube amp so I hadn't thought about charging the cap with the meter). Speaker IS still alive thankfully.
I may try removing the cap and see if that helps the turn-on thump. The thump isn't too bad, so it's not very important in any case. Would that cap also influence the turn-off thump? I would think a larger cap here would reduce the turn-off thump (by delaying the bias voltage reduction?).
Finally, one other issue I forgot to mention is that one channel remains silent for several minutes after the amp is turned on. The other channel will play immediately. I've been over the circuit and see no obvious problems. My soldering tip was getting cruddy and I did have some trouble getting good joints, so I suspect a cold joint somewhere. But are there other reasons this might happen?
Thanks again,
Carl
The voltage on the outputs was, indeed, measured without a load using a Fluke 12B meter (which I wouldn't do with tube amp so I hadn't thought about charging the cap with the meter). Speaker IS still alive thankfully.
I may try removing the cap and see if that helps the turn-on thump. The thump isn't too bad, so it's not very important in any case. Would that cap also influence the turn-off thump? I would think a larger cap here would reduce the turn-off thump (by delaying the bias voltage reduction?).
Finally, one other issue I forgot to mention is that one channel remains silent for several minutes after the amp is turned on. The other channel will play immediately. I've been over the circuit and see no obvious problems. My soldering tip was getting cruddy and I did have some trouble getting good joints, so I suspect a cold joint somewhere. But are there other reasons this might happen?
Thanks again,
Carl
The delay could be that the one channel's 33 uf cap is significantly bigger than 33 uf or the pot is significantly greater than 100k. Did you get 330 uf or 500K in there?
I was going to say pull the 33 uf cap to see if that speeds turn on, but the master has suggested otherwise. The cap helps quiet the bias reference voltage, removing psu ripple.
I was going to say pull the 33 uf cap to see if that speeds turn on, but the master has suggested otherwise. The cap helps quiet the bias reference voltage, removing psu ripple.
Thanks Nelson and Bob. I've been meaning to play around with relays for delaying B+ on tubes, so using one to short the output during startup on this amp might be a good way to begin. That's what I'll do next.
As for the one channel, I'll take a look but I'm pretty certain it's not any of the caps. The silence is much longer than a cap charging (at least, any that I've played with...). I haven't looked at the left channel problem methodically yet. I've usually fiddled with the inputs and outputs and eventually it works, but I've typically had the input switched off while doing so, so I don't know if that's what makes the channel work. Once it's on, it hasn't yet cut out, making it seem heat related or something like that (again, could be a cap but I don't think so). Maybe a cold joint that expands when heated up, making good contact? I'll probably re-solder those inputs and outputs and see if that helps.
Carl
As for the one channel, I'll take a look but I'm pretty certain it's not any of the caps. The silence is much longer than a cap charging (at least, any that I've played with...). I haven't looked at the left channel problem methodically yet. I've usually fiddled with the inputs and outputs and eventually it works, but I've typically had the input switched off while doing so, so I don't know if that's what makes the channel work. Once it's on, it hasn't yet cut out, making it seem heat related or something like that (again, could be a cap but I don't think so). Maybe a cold joint that expands when heated up, making good contact? I'll probably re-solder those inputs and outputs and see if that helps.
Carl
Clip your voltmeter to the gate of the FET and watch the DC voltage when you power on. How long does it take to get to your steady state bias voltage? If you used 330 uf and ended up with a megohm pot instead of 100K, it could easily take a couple minutes to come up.
If the Fet isn't conducting, there is nothing to cause heat that might make a bad joint suddenly conduct. How doe sit sound as it comes up? Slowly coming on or suddenly sound is there? Still it would be a good idea to reheat the connections, especially the pot.
If the Fet isn't conducting, there is nothing to cause heat that might make a bad joint suddenly conduct. How doe sit sound as it comes up? Slowly coming on or suddenly sound is there? Still it would be a good idea to reheat the connections, especially the pot.
OK so here's an update.
I measured with speakers attached (8 ohm +/-) and found NO DC voltage. All is OK there, thanks. The B+ and bias come up quickly on both channels so no problem with component values. But still I get no signal out of the left channel. I get a small AC (music input) signal just shy of the pot, but no signal on the slider side of the pot (the adjustable side of the pot). I tried re-soldering, but no go. While the pot measures OK, everything points to the POT being the problem, so I'm replacing it with a slightly better pot which I know to be working. Will report back when I'm done with this round.
Thanks again for all the help. I will try out removing the pot fully at a later date when my preamp (Bottlehead Quickie) has been upgraded. While I'm curious how that will sound, I want to nail down the circuit before I go on.
carl
I measured with speakers attached (8 ohm +/-) and found NO DC voltage. All is OK there, thanks. The B+ and bias come up quickly on both channels so no problem with component values. But still I get no signal out of the left channel. I get a small AC (music input) signal just shy of the pot, but no signal on the slider side of the pot (the adjustable side of the pot). I tried re-soldering, but no go. While the pot measures OK, everything points to the POT being the problem, so I'm replacing it with a slightly better pot which I know to be working. Will report back when I'm done with this round.
Thanks again for all the help. I will try out removing the pot fully at a later date when my preamp (Bottlehead Quickie) has been upgraded. While I'm curious how that will sound, I want to nail down the circuit before I go on.
carl
New pot is in and the amp has been playing on both channels most of the time. I have had a problem once so far with the left channel again, and it was with a phono pre. I played with the inputs on that and got pops and crackles, and eventually it worked, so I'm guessing it's now the phono (either cables or preamp). I'm hopeful the channel issue for the amp is fixed since I haven't had a problem using a CDP signal.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.