Fixed bias el84 jcm800

[AlNewman]

So I completed my first "real build" using ideas from Sluckey's November schematics, pairing it with a 12at7 PI tube, a presence control loosely ripped off from an HRD, and a preamp "borrowed" from a jcm 800. 

Added an effects loop, after the master volume, not in the schematic. 

It's a cool amp, bright and crisp.  Lots of headroom with the preamp down, lots of low volume crunch with the pre amp pushed, and a monster when everything climbs past noon.

Effects loop works good, I was a little worried about that.

I can see why everybody is working with turret and eyelet boards, because the point to point is basically a pain in the ass.  It does sound good, and it was an opportunity to try different things without commiting to a certain circuit.  If I found something that I would want to make 100 of, I'd definitely do it differently.

[Ronquest]

I'm surprised no one said anything about your switching the neutral and standby on the center tap.  You may want to change that.  On-off switch on the hot and move the standby to after the first reservoir cap.

[mresistor]

Are the big radial electrolytic capacitors glued down to the chassis?  If not what's to stop them for sagging down when the chassis is upright?

[AlNewman]

I don't see the advantage of switching the live?  Is there a purpose to that?  I chose neutral to keep cleaner connections, but it wouldn't be a big deal to change it if there was a solid reasoning.

The center tap is an acceptable route for the stand by, from what I've read.  What is the advantage of running it off the B+?

The radials are glued down, but they actually do make a very tight connection the way they're wired to the terminal strips.  I also went back and glued the long wire runs as well.

[sluckey]

I don't see the advantage of switching the live?  Is there a purpose to that?  I chose neutral to keep cleaner connections, but it wouldn't be a big deal to change it if there was a solid reasoning.

The idea of never switching the neutral is a carry over from the electrial industry. It's a safety code controlled by the NEC (as well as most state and city codes). The NEC (electrician's bible) states the neutral conductor SHALL be a continous circuit from the distribution panel to all devices. It SHALL NOT be interupted by switches, fuses, breakers, or any other device. It's a good safety rule. However, those building codes do not govern the wiring inside your amp.

By switching the neutral you allow 120VAC to still be present in the amp, through the power cord, fuse, PT primary, and all the way to the power switch. By switching the line the 120VAC stops at the power switch when off and there will be no 120VAC floating around in the chassis past the power switch. That sounds like a good idea to me.

The center tap is an acceptable route for the stand by, from what I've read.  What is the advantage of running it off the B+?

Try this simple test... Connect your voltmeter to the first bias cap (the one connected to the diode). What is the voltage reading with the switch in operate mode? What is the voltage reading with the switch in standby mode?    Still think it's an acceptable route for the STBY switch?

EDIT... Leave the meter connected for two minutes when in STBY mode.

[WimWalther]

Try this simple test... Connect your voltmeter to the first bias cap (the one connected to the diode). What is the voltage reading with the switch in operate mode? What is the voltage reading with the switch in standby mode?

With the amp cold, connect a voltmeter to the output of the bias supply. Then switch-on and watch how long it takes for the supply to come up to full voltage.

That delay isn't really an issue when all of the heaters are still cold, but switching the CT when everything is hot & ready to conduct means starting up with no grid bias.

This is not a desirable situation,  and if the fuse is spec'd right, it may well blow..

This is the same reason you don't want a bunch of extra filter capacitance on the bias supply.

ETA: Struck out errant comments.

[sluckey]

...switching the CT when everything is hot & ready to conduct means starting up with no grid bias.

That's not the issue at all. I'm guessing you have never done this little test. You might think the bias voltage goes to zero when you open the CT but that's not so. The bias voltage slowly rises to a very high negative voltage and that could be dangerous to the bias filter caps if the voltage rating is not high enough.

[WimWalther]

You're right, I'd never needed to try it as I'd never implement standby by knocking off the CT. In fact, I don't see much use for standby on most conventional designs for that matter.

It is necessary on the OTLs I'm used to working with, though.. for a couple of very good reasons.

But tell me, if knocking off the CT is effective at killing the current that charges the B+ supply, how does it still manage to charge the bias supply far beyond design intention? Where is the other phase of the PT sec. finding a ground reference? Leakage current through the PS filters? Bleeder resistors? Parasitic cap. to chassis?

ETA: Also, why don't the main B+ caps also attempt charge up to 2X voltage as well? It's a high-Z current in both cases, though, isn't it?

Sorry, this is totally counter-intuitive to me.

[mresistor]

The radials are glued down, but they actually do make a very tight connection the way they're wired to the terminal strips.  I also went back and glued the long wire runs as well.

Tight doesn't mean there wouldn't still be stress on the leads.
May I suggest the use of these, they're cheap and effective and unlike glue you can undo them and they're not messy like glue. I think they are avcailable in black to and also different sizes.


https://www.ebay.com/itm/355059604381?var=624327147981

[AlNewman]

Ahh, I never even thought of the bias.  I have my tv apart right now, so no room to try testing, but luckily I think I have 160v bias caps in that position, because that's all I had available.  I wouldn't even have put in a standby, except for the cathode follower.  Thanks for pointing that out Ronquest.

Those tie downs look handy, they'd work pretty good for some situations. 

[mresistor]

Those tie downs look handy, they'd work pretty good for some situations.

They stick like a nickname in high school too.

[sluckey]

Here's my Phoenix amp on the bench right now. The effects on B+ and bias voltages are shown on the schematic. I have known about the effect on the bias voltage for a long time but had never even thought to check the B+. You can see that the bias voltage nearly doubles and the B+ only drops to about 60V. I'm sure these voltages will differ depending on a particular circuit. I highlighted the sneak path that makes all this possible.

Anyhow, I consider these issues enough to prevent me from switching (or fusing) the PT CT when the circuit looks like this. This issue does not apply to cathode biased amps, or bias circuits with 150V rated bias caps, or any circuit that has a totally separate bias winding on the PT.

[AlNewman]

Yes, I'll definitely check it out next time I'm in there.

I do know my reservoir cap was at about 90v with the standby open.  My PT is 320v/side ac.  But I also don't have a bleeding resistor....Which maybe I should, but anyways.   It is definitely a good heads up.  To be honest, I'd rather leave it like it is due to how I have the layout set up, and if it isn't on a path to detonation, then that's probably what I'll do.  But I do want to check it at idle and start up to make sure there isn't too much stress on the caps or tubes. 

The neutral at the switch I will change, because I wasn't aware the electricians had a bible that said so.  I do like to practice due diligence. 

[WimWalther]

Here's my Phoenix amp on the bench right now. The effects on B+ and bias voltages are shown on the schematic. I have known about the effect on the bias voltage for a long time but had never even thought to check the B+. You can see that the bias voltage nearly doubles and the B+ only drops to about 60V.

Ok, so the return path is through a bleeder resistor (if present..), though in absence of such it may also *possibly* find a path via current leakage through the main filter caps.. maybe? Guess it depends on how high the voltages get.

This issue does not apply to (...) any circuit that has a totally separate bias winding on the PT.

Oh.. really? Have you actually tested this as well? If the bias is charging to ~2X in the case already discussed, what would prevent it charging to a higher voltage derived from 1/2 of the PT sec. + the voltage of the bias tap?

ETA: Oops, you said "totally separate". That's a different situation, then, though I don't think I've ever seen such a design.

Now I'm really wishing I had an amp to test this with, like.. the Eico ST70, which has a separate bias tap.

Interesting and instructive stuff, for sure.

[sluckey]

Ok, so the return path is through a bleeder resistor (if present..), though in absence of such it may also *possibly* find a path via current leakage through the main filter caps.. maybe? Guess it depends on how high the voltages get.

It behaves the same without the bleeder, although the voltage is slightly different. I just highlighted the bleeder resistor because it's easier to visualize the sneak path.

This issue does not apply to (...) any circuit that has a totally separate bias winding on the PT.

Oh.. really? Have you actually tested this as well? If the bias is charging to ~2X in the case already discussed, what would prevent it charging to a higher voltage derived from 1/2 of the PT sec. + the voltage of the bias tap?

Duh! I said separate bias WINDING, not a tap on the HT winding.

[WimWalther]

Duh! I said separate bias WINDING, not a tap on the HT winding.

You must have missed that I quickly caught my misunderstanding and came back to add:

ETA: Oops, you said "totally separate". That's a different situation, then, though I don't think I've ever seen such a design.

[sluckey]

You must have missed that I quickly caught my misunderstanding and came back to add:

That edit was not there when I began my reply.

[AlNewman]

Well, in testing, I've discovered that with standby open voltage at the first bias cap sits at about -106V, with voltage at the grid of the power tubes at around -25, (normally -14V).  All B+ voltages vary between 90V-70V.  All the little tubes conduct minimally with standby open, and the power tubes don't.

Upon start up, the current through the power tube cathodes climb gradually from 0 - operating voltage, while the plate voltage spikes at around 375V and gradually drops to operating voltage.  Seems to me to be a soft start up for the power tubes, as well as the rest of the tubes and capacitors.... Maybe I'm missing something, but I'm having a hard time seeing the disadvantage to this system. 

One concern I have is the 120k 1st bias resistor runs fairly hot, at about .5 watts when in standby.  It's only a 1 watt resistor, but I do have some 2W 100k, which I'll pair with a 1W 68k for the 2nd resistor, and that should get me in the ballpark again.  I think this should be a pretty solid circuit with 160V caps, I'm still open to feedback though.

[WimWalther]

Well, in testing, I've discovered that with standby open voltage at the first bias cap sits at about -106V, with voltage at the grid of the power tubes at around -25, (normally -14V).  All B+ voltages vary between 90V-70V.  All the little tubes conduct minimally with standby open, and the power tubes don't.

Upon start up, the current through the power tube cathodes climb gradually from 0 - operating voltage, while the plate voltage spikes at around 375V and gradually drops to operating voltage.  Seems to me to be a soft start up for the power tubes, as well as the rest of the tubes and capacitors.... Maybe I'm missing something, but I'm having a hard time seeing the disadvantage to this system. 

One concern I have is the 120k 1st bias resistor runs fairly hot, at about .5 watts when in standby.  It's only a 1 watt resistor, but I do have some 2W 100k, which I'll pair with a 1W 68k for the 2nd resistor, and that should get me in the ballpark again.  I think this should be a pretty solid circuit with 160V caps, I'm still open to feedback though.

With 160V caps in the bias supply, they're safe. Your bias dropping resistors are likewise at safe power levels, though subbing higher-rated parts is never a bad idea in that situation.

At -25V Eg1 the power tubes must be approaching cutoff, or at least running very slowly as evidenced by the initially high B+. I'm sure it does produce a fairly soft-start.

Can't really argue with your conclusions. It does seem like a reasonable and safe arrangement, at least so far..