SE EL34 Plexi-verb

[martin manning]

Pin 4 is ground, so there is no reason you can't use that for R37's grounded end. The series R36 and R37 (total 20k) are paralleled with the series R31 and R32, so I would expect to get only about 1M measuring either R31 or R32- they are only 20k away from being paralleled with one another.

[Phil_S]

That's good news because the board is soldered and tested with the meter for continuity. It is hard to work on those little boards with all the little holes, etc. It was good I had a #1 pointy tip for my iron that's been lying around for, what, maybe 10 years! Making a good solder joint on the board was not so easy for me. One cool thing I realized is the narrow strip between the copper traces won't attract any solder and in a pinch it is possible to run the pointed iron down the strip to break an unintended bridge from this rookie's sloppy work.

I'm still waiting for the 4700uF cap for the power supply board. It should be here in the next day or so. Once that's built and I know it's 3-D envelope size, I can make sure it fits in the chassis and start cutting and drilling. It's all going to be a tight fit.

[Phil_S]

I've been busy...made my fair share of mistakes, do overs, didn't order all the parts...nearly done with the amp part, but may not get to fire it up 'til next week. Next week, some warm days predicated. Might be able to get outside to cut wood for a cab, not sure...

I should note that noval sockets are wired for a pair of Mullard 6EU7's that I've got, not 12AX7's.

[Phil_S]

more...

[Phil_S]

I had a few minutes of fun time this morning but need to move on to other things. I started with the light bulb limiter. I removed the Belton BTDR-3 brick to keep it safe from any risk.

With the 25W bulb there was nothing noticeable, but that is pretty low.

With the 40W bulb, I noticed the 5/15VDC supply board for the reverb did not have any output on the 5V supply and it was considerably less than what I expected with the 40W bulb in line. I shrugged because I don't have solid state experience and figured the parts on that board were cheap enough. My general observation was that HT VAC was about 35-40% of what's expected without the limiter. I reasoned I should be seeing 1-2V on the 5V output and 5-6V on the 15V output. When I saw just 0.18V on the 15 output, I should have stopped there. But no, I forged ahead.

When I lit up the 60W bulb, I let the smoke out of the 5/15V supply board and immediately shut it off. At that point, I removed the offending board for examination. This will allow me to continue with the rest of the amp. It just renders the reverb non-operational.



For the record, the 60W bulb flashed bright very briefly as the smoke emerged, too quick for anyone but Superman to move quick enough to the off switch.

I will do my best to figure out that went wrong with that board. My immediate suspicion is that something was grounded incorrectly. In any case, I'm likely to trash the board but I may salvage some parts depending on what I learn. I will look for something besides strip board for the power supply but don't know enough about this to imagine what I might find.

[martin manning]

Rats, I was looking forward to seeing how this choke input power supply turned out. I think you were on the right path removing the reverb module. Better to isolate parts of the system and test individually, IMO. Here I would test high and low voltage supplies separately. Which parts were smoked?

[Phil_S]

Martin,

Rats? Heck, I'm really disappointed I have to deal with that power supply board. The rest of the amp should still work.

I think choke input is still on line. It is only the power supply board for the reverb that is now off line. I was not recording voltage while the limiter was in use. If I recall, with the 40W bulb, HT was about 375VAC (roughly half of expected; that's across both legs) and B+ 290 after the choke. Those are without tubes. I am convinced we will get a good result on the B+.

I am not sure what smoked on the 5/15V board. I see one cap that has some grey where the white stripe negative indicator is on the outer coating. I wonder if I had that one installed backwards. I need to go over the whole thing, so I need to find time to do that.

My first real order of business will be to reduce the filament voltage. I am expecting 7.5V unloaded, but I don't know what it will be with tubes. I need to see that to determine whether/how to deal with it. I was not paying attention and then I realized I forgot to stop and test after wiring the heater harness. I don't want to fry the preamp Mullards. After that, I'll be looking to record B+. Sit tight. Next week sometime.

Phil

[Phil_S]

Since there were no posts after mine, I'm editing the post and replacing the schematic. If you viewed it earlier, you aren't nuts as it has changed.

I must have a bad solder somewhere. I had the chassis standing on the PT end and got bad voltage readings. I found something to stand it on and placed the open side up. Now I have a full set of voltage readings to present. Preamp voltages are too low and I will have to fiddle with the dropping resistors in the supply ladder.

I wouldn't mind hearing some opinions on where you think the EL34 screen voltage ought to land or if I should just leave the 5K dropping resistor alone.

I'm perplexed about V2. The voltage drop across the 100K tells me it's pulling nearly 11mA and that seems rather excessive for a 6EU7 (same spec as 12AX7). I'm not quite sure what to do about it and why I've only got 9V on the 1st plate. This seems all wrong and I'd be grateful for a suggestion. I suppose I should swap the tube and I will later.

Of special interest is the choke input. The power tube came in very close to what we discussed in the other thread. Martin was using B+ 340 for the plate supply and that's right where it ended up.

The heaters were at 7.17VAC and I tamed that to 6.45 with a 0.25Ω resistor between the PT and one leg.

An aside on the 5/15V supply board for the reverb, there is a blown diode. I don't know if anything else went up in smoke. Naturally, I hope not. I'm going to rebuild on a different type of board. I'm convinced it was not wired correctly.

[martin manning]

Looks like you have slipped a decimal point on many of the current calcs... you're not using a slide rule are you? ;^)

I see 7.7mA through R21, and 7.4 through R29, so only 0.3 going to V3, confirmed by the drop across R18. The 7.4mA through R29 doesn't seem to square with the currents through the first four triodes though- only 0.54mA summing up the cathode currents. Looks like the excess is flowing into the area of the second two triodes.

Screen voltage was supposed to come out around 250, so it's close. The excess current draw around V2 is dragging it down though, so sort that out first.

[Phil_S]

What's a factor of a tenth among friends? LOL, I was not paying attention. This is making a whole lot more sense.

I am inclined to think the problem with V2 is a wrong resistor value on the cathode follower. The CF should have high cathode voltage and it does not. The gain side of that tube has ridiculously low plate voltage. None of that tube is making sense.

I know that I need to sort out the V2 problem first. After that, I am inclined to change the three 1K dropping resistors to 1K to up the voltages in the preamp and I might swap the 5K for a 1K also. I have a bag of 100 1K 4W, so that is the only immediate choice, though I do have some odds and ends in higher wattage, including one 2.7K. We'll see. I think I'm getting ahead of myself here. Addressing V2 might surprise me with voltages all around.

The overriding good news is that probing the tip of the input jack pops at the speaker, so I know the amp is passing a signal all the way through!

Edit: found the problem, and it was as suspected. I managed to switch the cathodes on V2, so the follower was into 1K and the gain stage was into 100K. D'oh! I am going to write this off to not paying attention when I decided to switch from 12AX7 to 6EU7. Now I have to turn the volume nearly all the way down on the bench or the pops are jump out of your skin loud. Voltages have come up to where they are reasonable.

I am thinking higher preamp voltage might be nice to give a bit more clean headroom. I don't think it will matter much whether I change the 3x 10K or not as that part of the amp is drawing so little current. The questions are whether to reduce R27 and R26 to set up a higher voltage at the front of the line. Do you think 2.7K and 1K respectively will give me more to work with? Where I'm headed is that the CF on the 2204 should be at around 150V. I don't feel compelled to get that high (and am pretty sure I can't get there from here) but would like to higher than 80V.

Probably the B+ tweaks should wait for me to rebuild the 5/15V board for the reverb, as I'd like to see that come on line.

[Phil_S]

The second 5/15V board gave up the smoke. It was the same scenario. It was fine with the 40W bulb inline. It blew when I upped the limiter to the 60W bulb (which is getting close to full power.) the 60W bulb went bright very briefly, too quick to act and then...poof!

I did a minor redesign, see schematic, as I thought maybe I didn't understand what I was doing with the earlier one. There is an error on the schematic that I did not make in real life. The two 2.5V feeds are not shorted to each other. The diode goes directly from the ground to the upper input voltage source.

This time there are 2 smoked diodes -- the two upper ones. I am 100% positive there is something wrong with what I'm doing, but I don't know what. I've willing to keep trying, but am not willing to engage in additional acts of futility. Repeating what I've done seems futile. Any help here would be appreciated.

[martin manning]

In your schematic just above the cap connected directly to the upper AC leg is backwards and probably shorted now. It is correct in the whole amp schematic. And of course the dot on the lower AC leg (wire to the cathode of the diode going to ground) shouldn't be there as its shorting the AC.

[Phil_S]

In your schematic just above the cap connected directly to the upper AC leg is backwards and probably shorted now. It is correct in the whole amp schematic. And of course the dot on the lower AC leg (wire to the cathode of the diode going to ground) shouldn't be there as its shorting the AC.

Bada bing! The cap is backwards! Amateur hour ;-{ I have a feeling I'm a slow learner and did this twice.

The dot on the lower AC leg is just something Jschem did and I was too lazy to undo it. That one was wired correctly.

I'll swap out the cap and the two diodes and see what happens. Like this...

[JazzGuitarGimp]

I probably should stay out of this one because I have no experience with voltage multiplier circuits (that is what this is, right?) But I am inclined to ask, does the transformer winding have a center tap? I ask this because you have labeled the ends of the winding as 2.5VAC each, which, to me, implies a center tap. Anyway, if it does have a CT, I am pretty sure you want to leave it floating, rather than ground it. There, that's all I'm sayin'! Good luck!

[Phil_S]

Hello Lou. Thanks for weighing in. The winding is a 5V rectifier winding, no CT. I think I finally uncovered the real problem. The real problem is that I am a novice at using strip board and, worse, I found out I'm not as smart as I hoped. This is embarrassing to admit, but I'm willing. In addition to putting the cap in backwards, I did much worse. I put the cap and one diode on a single strip without cutting the trace -- the trace needed to be cut in two places. Basically, I shorted the cap and the diode. D'oh! So, I jumped one leg of each over an unused strip for a proper circuit path and I think it will work now. It was Martin's observation about the cap being in backwards that prompted me to "see" instead of just "look." I'm a little gun shy to fire it up, but it is installed and so I'll have to get on with it soon. If it goes poof again, I'll go around yet again. Eventually, this will work.

Today, I am planning to build a cab for it while I have the weather on my side, as I have no indoor space to make sawdust. It will be a push to get it finished.