• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Kofi Annan in: "Tube Amp for Multi-Way Speakers"

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've always connected the grounds from the power stage, preamp stage and input grounds to the chassis near the first PSU cap, then connected that chassis-grounded star to the chassis safety ground next to the IEC inlet along with the earth connection. Two chassis connections.

I've never had a problem with hum, but now that I have a preamp in the mix things may change.

Are there safety concerns if I don't attach the preamp / power / audio grounds to the chassis?

Kofi
 
Last edited:
To clarify, I ground the loop breaker to the chassis using a different connection point than the AC safety ground. The safety ground has it's own connection.

It's good that you have not experienced any previous hum problems, however there should be only one connection point from signal ground to chassis ground. Whether you use a loop breaker or not, it is possible to develop a ground loop within the chassis itself which may show up as hum.

Yes, there are safety concerns if signal and chassis ground are not connected. Some type of catastrophic component failure could put HV on the signal ground. Also, grounding the chassis minimizes noise and hum.

I hope this is helpful.
 
Scott,

Yes! Thanks for posting. For now, I'll stick with my method and ensure the master star is attached to the chassis.

Also, after further examination it appears that I'm running the EL84s past max dissipation and voltage (not that much, but still...). I should probably drop the B+ to 300V with a resistor.

I know that other designs run EL84s hotter, but I'd be concerned about long-term reliability and general failure.

Am I being too concerned here?

Kofi
 
I've seen other designs as well that run the plate voltage higher in pentode with the screen at a lower voltage. I don't have a lot of experience, any actually, with the EL84 in parallel push-pull pentode, but I remember reading on this forum somewhere that original production EL84/6BQ5 were better at handling the higher voltage with a well regulated screen.
I'm guessing you will be using new production tubes. If it was me, I'd stay within the published specs if you are looking for long-term reliability.

What is the final configuration for your new amp? Pentode wired cathode bias? Individual cathode resistors?
 
Here's the final design. Note that the voltage is 325 in the schematic, not 300. Reducing the voltage will negligibly increase distortion and reduce output current, but would pull the EL84s back in line with max operating voltage and power dissipation.

For the record, I'll be using some NOS Russian mil spec equivalents (matched octal) that should be rugged enough for the task, however it's possible that the client changes does some tube rolling in the future with less durable models and winds up red plating.

I do currently run a Baby Huey EL84PP with about 320 on the plates, but since this will wind up with a client, I think it would be wise to ratchet it back a bit.

Kofi
 

Attachments

  • EL84PPP FINAL.PNG
    EL84PPP FINAL.PNG
    223.4 KB · Views: 195
That's going to be a nice amp. That's a great selection for the OT. If I was going to keep it I'd leave it like it is power supply wise. I mean it's right on the edge after taking into account the cathode voltage. But as you say, you can't tell what the client may do. I'd definitely not make any changes to the power supply until you have it powered up and see where you actually are.
 
Last edited:
Thanks! You should know that the design was a team effort (kevinkr, Lingwendil, rongon and others with rongon working up the final design in LTSpice).

I'm hoping to complete the wiring by tomorrow sometime. I'll post after the initial power-up with a list of all the components I blew up and a request for a local therapist's contact info.

Kofi
 
OK, so I powered up tonight with a variac and at about 60VDC out the PSU smoked two 20W resistors in series and popped the fuse. I disconnected the PSU from the circuit and ran a 1.2K 20W resistor across it, which it promptly smoked at around 70 VDC.

If there was a short in the supply, it should have popped the fuse but not fried the resistor, so I'm wondering if there's something up with the transformer. I wired the red leads together and the black leads together on the primary side (wired in parallel) for 115V and grounded the purple wire, which I believe is for the screen.

I'm about out of brain cells for the evening, but any advice would be much appreciated.

Thanks,
Kofi
 
Administrator
Joined 2004
Paid Member
Are you sure those 20W resistors are actually 1.2K? At 70VDC the resistor would be dissipating just 4W, consequently it seems unlikely that that is the resistor value.

But I wonder why you are doing this at all?

I would normally apply power to the entire amplifier and measure all voltages and currents starting at a small fraction of the rated line voltage. You can also use a ballast lamp in series with the primary to minimize mischief in the event of a mistake.
 
Are you sure those 20W resistors are actually 1.2K? At 70VDC the resistor would be dissipating just 4W, consequently it seems unlikely that that is the resistor value.

But I wonder why you are doing this at all?

I would normally apply power to the entire amplifier and measure all voltages and currents starting at a small fraction of the rated line voltage. You can also use a ballast lamp in series with the primary to minimize mischief in the event of a mistake.

Thanks for the reply.

So, I Did bring up the voltage slowly and found that the voltage dropping resistors started smoking about 70VDC in. That’s when I tested the supply across the 1.2k and it burned out at around the same voltage. I’ve probably got a little fatigue and I’m not thinking clearly, but I’m not understanding why the resistors would fry in these circumstances.

I’ll try again tomorrow. Any clear, non-exhausted insights would be welcome.

Kofi
 
I did. Just to prove that I hadn't discovered a loophole in Ohm's Law, I tried again with a 400R 10W job and was able to get about 60VDC across it without letting the smoke. The resistr was labeled 12W, but I pulled it from an old amp and it may have seen better days.

The fact that I smoked the two 75R resistors in series with the circuit implies a short somewhere. I also noticed that one of the screen grid windings has about half the resistance to ground as the others. I'm a little concerned that it implies a shorted turn somewhere.

More testing to come.

Kofi
 
So the latest test with the 400R-10W is with the amplifier circuit connected?

Do you have a power supply schematic handy that you could post?
Have you tried powering just the supply with 120VAC and see what the output voltage is? If there is problem, with the supply you will blow the main fuse.
 
So the latest test with the 400R-10W is with the amplifier circuit connected?

Nope. Just across the resistor. Wanted to pull the circuit and test separately, which, admittedly, is not conclusive. I think I just wanted to make sure I was getting rectified DC.

Do you have a power supply schematic handy that you could post?
Have you tried powering just the supply with 120VAC and see what the output voltage is? If there is problem, with the supply you will blow the main fuse.

I have attached the PSU design. Note the 80R resistor was in series with the circuit but without the final cap. And it smoked at around 50VDC.

Thanks so much for all the help!

Kofi
 

Attachments

  • image.png
    image.png
    166.6 KB · Views: 129
With 60VDC across a 400R resistor, it is dissipating 9W, so yes, it will get very hot and I believe de-rate quickly from there. 65VDC and you are over the limit of 10W.

I modeled your same power supply in PSUD2 and got the same exact results, of course. I'm kind of at a loss.

Did you measure the AC transformer output voltage with the secondaries in parallel?
 
Did you measure the AC transformer output voltage with the secondaries in parallel?

Do you mean pull the AC from the circuit and measure the full output voltage from the toroid? Nope. Haven't done that. I'll give that a go and post results.

My paranoia is kicking in here regarding a resistance measurement. One of the left channel screen taps is measuring about 60R to ground while the other three are measuring about 140R to ground. Wondering if there's a shorted turn in the transformer that may be causing this?

Kofi
 
I don't think you should read resistance that low from any of the output xfmr primary taps to ground including the screen taps. You would get a high charging reading through the power supply caps. But if your power supply is disconnected from the amp circuit, you should not read any resistance from the taps to ground.
 
Yes, somewhere. That means if you were to measure resistance from where your B+ connects to the OT, and ground, (with the PS disconnected), you would read about 100 ohms. That's a problem.

You said you were getting strange resistance readings on the OTs, why not try disconnecting the OT primary completely from the amp circuit and see what you get. Also disconnect one end of R24 feeding B+ to your preamp/splitter section just to remove that from the equation as well.

Nothing in your final schematic indicates that you should read 100 ohms to ground.

No wonder your dropping resistors initially failed. If you indeed have 100 ohm load on the power supply, then even at 70VDC, it's trying to pull 700mA.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.