Reel to Reel Conversion

[Brendan]

Greetings!
A few months ago I found a Voice of Music reel to reel player at my local Goodwill and once I determined it had a single 6L6 and a handful of 12a_7 tubes, I grabbed it.  I know a SE 6L6 isn't considered optimal by some, but I want to give it a try.  This isn't my first time inside and amp, but it is my first build.  I did want a good challenge, so I made my own chassis from a 12"x12" piece of 22 gauge steel from Home Depot and used a cheapo Harbor Freight bending tool to put it into the proper shape.  I used a uni-bit and a HF bi-metal hole saws to make the tube socket holes. ( the next build will be all octal and I 'm buying a damn punch!)  I decided to start with the AA764 non-tremolo Champ schematic. I understood there were going to be tweaks needed if I used the 6L6. Pics!

The PT is a 400v-0-400v monster and when firing up for the first time with no tubes ( current limiter was attached), the readings at the rectifier socket looked good. I installed the 5Y3 rectifier that came with the donor and the B+ was too high.  590v! That's just with the rectifier installed.  The JJ can cap setup has a 500v limit. I removed the 5Y3 and gave it a gentle shake. There was something loose inside.  I tossed it and installed a 5y3GT I had at home.  Still with only the rectifier tube installed, voltage at the filter caps was now at 470v, 460v and 445v.  With a JJ 6L6GC and a 12AX7 installed, voltage dropped to 371v, 357v and 235v.  The voltage at that third filter cap for the preamp concerned me.  That puts about 139v at pin 1 for  preamp plate. I was expecting over the factory spec of 210v or so.  I confirmed resistors were correct for the schematic and everything is connected properly. I know that's a king kong PT for a champ build, but I don't think that's a problem for the JJ 6L6CG. What troubles me is the 235v reading on that third filter cap. So here's the question - could the 590v burst on the first fireup have damaged that cap?  The amp does sound fairly quiet at idle and I did plug in and play a guitar.  Virtually no gain though and not much volume. I attribute that to the low voltage at the filter cap. Is that a reasonable assumption? I have a new one on order as well as new resistors.

Another note - the light bulb current limiter ( 150w) only lit up once and that was when I accidentally left my DMM in capacitance testing mode.  I'm not sure if that's relevant.

Here are my last readings:
B+ 371, 357.7, 235

6L6GC
3) 355.8
4)357
5)+.021
8)27.96

12ax7
1)139.1
2)0
3)1.156
6)161.6
7)0
8)1.16

I know I have to make adjustments for the voltage increase and tube change, but is there anything here that might point to another source for the low preamp voltage?

 

[Paul1453]

Could you post the schematic with the mods you made?

You did a very nice job on your chassis!

Just a guess, but maybe something isn't grounding well to your painted steel chassis?   

[Brendan]

Could you post the schematic with the mods you made?

You did a very nice job on your chassis!

Just a guess, but maybe something isn't grounding well to your painted steel chassis?   

Thanks, Paul. I'll double-check my grounds. I did bring all the grounds together to one central point near the transformer with the only exception being the power cord common wire that's mounted near the transformer as well. I used small round stickers to cover the parts of the chassis that I didn't want paint to get on, but it's a good idea to double check since I've only assumed it's good due to the quiet idle.

Here's the schematic with changes.  I pretty much stuck with mostly stock AA764 components except for the 34K resistor (since I'm using one input jack) and the two cathode capacitors that were bumped to 50v simply because I had those.  I expected the uf value to be most important. Is that correct? Otherwise I bumped up the wattage on a few resistors. The gut shot picture on my first post shows a Xicon 250 ohm resistor that's going to come out, so it's not on this schematic. I was testing to see how far I could bring voltage down.  I was entertaining a 6v6 option at one point, but that PT is too big and I don't have the real estate for a Zener farm.  I decided to move forward with the 6L6 since that is what that transformer came with .

[Paul1453]

Well 1st your schematic shows you should have +330 at your last filter cap, but you report +235 almost 100V low.

Double check that 10K 3W resistor.  If correct maybe lift the downstream leg feeding the 12AX7 and verify +330V.

Then you should have nearly the same voltage on the plates of your 12AX7, which you don't.
Maybe you have some leaky caps?  Or a bad 12AX7?

[Willabe]

Paul, that's Fenders schematic. He changed/added what's in red.

[John]

Try putting a smaller bulb in your limiter, maybe a 60W. If you're drawing a good amount of current (to ground) that should glow pretty good. Like Paul says, try lifting the wire at your term strip downstream. With no current draw that should be reading 300/whatever. If so, then the problem is after the connection on the term strip. If it's still low there (basically your node C) then I'd suspect the C node on the cap can.


Probe carefully. ;)

[Brendan]

Sorry for the confusion, I'll make revisions to the schematic.

I did confirm the resistor, but I'll lift that wire at the terminal strip and see what changes.  I'll pick up a 60w bulb before I do that.

Thanks for the input!

[Paul1453]

Pin 8 on the AX7 also appears to have a 2.2uF cap.
I don't think that is your problem, but I think that might be an error on the schematic.
I think that cap maybe should be 25uF like the others.   

I visually confirmed the color code on that resistor, looks like 10K to me.

If you just snip that red wire at the cap can you can check that voltage pretty safely.
And then just solder it back on again.
If you do that, remember to discharge that cap before reconnecting the wire so you don't get bit. 

[PRR]

First oddity: you have two halves of one 12AX7, with identical bias resistors (the 47r was picked to be nearly-negligible); yet you have very different plate voltages. I would expect 10V difference. 22V difference *may* just be an old or mis-made 12AX7, but it provokes wonder.

BIG oddity: the "10K" shows 357V one side, 235V the other side, 122V difference. 122V across 10K is 12.2mA. But we know two-halves 12AX7 is 2mA (more or less), NOT 12mA.

So....

The "10K" is not 10K, or...

There's a big leak somewhere. The only un-accounted branch here is the capacitor. After the first few seconds, it should be down to under 1mA, decreasing to zero by the time you get a meter in there.

10mA of leakage at 235V is 2.5 Watts. Some real heat. CAREFUL finger pointing, maybe just sniffing, may find a hot-spot. If it is inside the cap, the outside may take time to warm, but that's enough heat to eventually blow-up a small/medium capacitor.

[Brendan]

Pin 8 on the AX7 also appears to have a 2.2uF cap.
I don't think that is your problem, but I think that might be an error on the schematic.
I think that cap maybe should be 25uF like the others.   

I visually confirmed the color code on that resistor, looks like 10K to me.

If you just snip that red wire at the cap can you can check that voltage pretty safely.
And then just solder it back on again.
If you do that, remember to discharge that cap before reconnecting the wire so you don't get bit. 

Thanks for checking, Paul.  I just found the Triode Electronics schematic for their AA764 kit and they also show 2uf for that resistor too.  I did test the 10K resistor a few days ago and it does measure around 9.95K.  I plan on adding a permanent bleed resistor to this one!

[Brendan]

First oddity: you have two halves of one 12AX7, with identical bias resistors (the 47r was picked to be nearly-negligible); yet you have very different plate voltages. I would expect 10V difference. 22V difference *may* just be an old or mis-made 12AX7, but it provokes wonder.

BIG oddity: the "10K" shows 357V one side, 235V the other side, 122V difference. 122V across 10K is 12.2mA. But we know two-halves 12AX7 is 2mA (more or less), NOT 12mA.

So....

The "10K" is not 10K, or...

There's a big leak somewhere. The only un-accounted branch here is the capacitor. After the first few seconds, it should be down to under 1mA, decreasing to zero by the time you get a meter in there.

10mA of leakage at 235V is 2.5 Watts. Some real heat. CAREFUL finger pointing, maybe just sniffing, may find a hot-spot. If it is inside the cap, the outside may take time to warm, but that's enough heat to eventually blow-up a small/medium capacitor.

Thanks, PRR.   The 12ax7 is indeed old. I do have two more here from the donor, but they are a very clean pair of Mullards that I don't want to gamble with.   I already have a JJ replacement on its way for testing. I've had the amp on for as much as five minutes without any smells or odd noises.  I've been worried about the health of that can cap after that first incident, so now I've ordered a replacement for that as well. I'll feel better getting getting those out of the equation and they really don't cost much. Thanks for giving this a look-over. If I still have an issue, I'll try to isolate as has been suggested. I'll post my findings.

I really appreciate everyone's help. 

[sluckey]

You say the amp has "not much volume". What does that mean? Is it only as loud as two people talking at normal levels in a quiet room?  Would your wife or neighbors complain if you cranked it wide open? A properly working Champ is loud enough to annoy the neighbors if you live in a suburban community!

About your voltage readings... Was the amp still connected to your light bulb current limiter? If so, plug the amp straight into the wall and recheck all voltages.

I'm also concerned about the voltage on that third cap. But even though 235V seems wrong and the currents don't add up, 235 should still be enough to be a very loud amp. The cap is suspect, but you have a spare cap in that can. Move one end of the 10K over to the spare cap and just use it instead. Make a difference?

Looking at your input jack it appears to be wired wrong. The 33K that connects to the shielded wire should also connect to the tip lug of the jack, ie, the same lug that one end of the 1M resistor is connected to. Maybe it is but that is not plain to see in the pic. If the 33K is actually connected to the switch lug, the volume will be very low. A pic showing a different angle of the jack that shows all three lugs would confirm. Take a look.

[Paul1453]

10mA of leakage at 235V is 2.5 Watts. Some real heat. CAREFUL finger pointing, maybe just sniffing, may find a hot-spot. If it is inside the cap, the outside may take time to warm, but that's enough heat to eventually blow-up a small/medium capacitor.

So for a KISS Army amp tech like me, (Keep It Simple Stupid),
That cap can is going to get HOT.

So by extension, for KISS techs, can we say that whenever we notice an E-cap getting really hot that it must be leaking some voltage and we should replace it asap, before it blows?

[jjasilli]

Also, check plate dissipation.  I put a 6L6 in my champ & used an 880Ω cathode resistor for good bias.

[Brendan]

You say the amp has "not much volume". What does that mean? Is it only as loud as two people talking at normal levels in a quiet room?  Would your wife or neighbors complain if you cranked it wide open? A properly working Champ is loud enough to annoy the neighbors if you live in a suburban community!

It was as loud as two people talking. The nearest guitar had hot P90 pickups, but the amp was very clean when at full volume.

About your voltage readings... Was the amp still connected to your light bulb current limiter? If so, plug the amp straight into the wall and recheck all voltages.

I made that mistake earlier, but it was connected straight into the wall when I got these readings.

I'm also concerned about the voltage on that third cap. But even though 235V seems wrong and the currents don't add up, 235 should still be enough to be a very loud amp. The cap is suspect, but you have a spare cap in that can. Move one end of the 10K over to the spare cap and just use it instead. Make a difference?

I'll try that before I swap in the new can cap.

Looking at your input jack it appears to be wired wrong. The 33K that connects to the shielded wire should also connect to the tip lug of the jack, ie, the same lug that one end of the 1M resistor is connected to. Maybe it is but that is not plain to see in the pic. If the 33K is actually connected to the switch lug, the volume will be very low. A pic showing a different angle of the jack that shows all three lugs would confirm. Take a look.

You are correct! Thanks for spotting that.

[sluckey]

Notice that a lead of the 1M is used to jumper the ground lug and switch lug together. This grounds the jack when nothing is plugged and helps keep the amp quiet.

[eleventeen]

"So by extension, for KISS techs, can we say that whenever we notice an E-cap getting really hot that it must be leaking some voltage and we should replace it asap, before it blows?"


Sure. But a cap need not get hot for it to be defective. All you are doing replacing a "hot" cap before it blows is preventing it from spewing nasty stuff all over your amp's innards, which is a worthy cause. But it is only going to get hot when blowing its guts is imminent. IOW, you would have to be watching it and feeling it mere minutes or even seconds before it lets go.

[Platefire]

Brendan

Glad to see you finally posted your project here. Looks like your getting some good help. I just know when you get all the little gremlins out, it's gonna scream Platefire

[Paul1453]

So in an effort to regain my former Keep It Simple knowledge, I'm thinking of two specific situations to apply this:

1.  Any unknown gear.  Upon initial powering up, checking proper operation, voltages etc.
If during this short 2-5 minute powering up, after unplugging, I touch all my cap cans and E-caps and notice if any of them are hot.
These Hot cap cans or E-caps should be immediately replaced before any further testing.   

2.  If I have a properly functioning amp that all of a sudden displays marginal/poor performance.
I do the basic swap of current tubes with known good ones and problems continue.
Before going off in the weeds chasing snipes, I should:
Touch all my cap cans and E-caps and notice if any of them are hot.
Hot cap cans or E-caps should be immediately replaced before any further testing.   

I was amazed at how many times these Keep It Simple techniques would often fix the problem without any other real troubleshooting taking place.   

[John]

One other thing, which I'm sure is just camera angle. But on the second pot, your 250pf cap isn't touching the pot casing, right?

[Brendan]

Also, check plate dissipation.  I put a 6L6 in my champ & used an 880Ω cathode resistor for good bias.

Thank you.  I know I'll have to make those calculations, but do you recall what else you may have changed? I'm looking forward to getting this dialed in.

[Brendan]

One other thing, which I'm sure is just camera angle. But on the second pot, your 250pf cap isn't touching the pot casing, right?

Correct, that's not a good angle the the cap and leads have a nice gap.  I appreciate the extra eyes on it!

[Brendan]

Brendan

Glad to see you finally posted your project here. Looks like your getting some good help. I just know when you get all the little gremlins out, it's gonna scream Platefire

Thanks, I'm really looking forward to it.  I'm fixing the input jack now and will have the new can capacitor and 12ax7 here on Friday.  I could test the low voltage situation tonight with the spare capacitor in the can, but I want to test this with a known good 12ax7, so Friday it is. 

[Brendan]

My replacement parts arrived a bit late, but I got busy tonight and installed the new can cap.  I did the first startup with no tubes and checked the readings.  My DMM normally gives a momentary "OL" display and then shows the actual voltage.  The first test at a HV secondary with no rectifier gave a longer OL display and did not produce an AC voltage display as it normally would, so I shut it down. I tried a minute later, and I got 394VAC which is what I would expect.   I thought the first attempt was odd, but I went ahead and put the rectifier in and got 570V at B+.  I was expecting  470v .  I probably should have stopped there, but with a known good power tube and preamp tube in, I  did a quick check of B+ and got 700+V. I shut it down immediately.
All I did here was install the can cap and I triple-checked my work.

Thinking back to when I first fired up this amp, I had a high voltage reading that I thought was possibly due to a rectifier tube. That was over 550V.   I'm going to disconnect the PT secondary leads tomorrow and do more testing.  I'm not sure if a PT can have intermittent issues, but my first long "over limit" warning from the PT tonight may be indicating so.  If that's the case, I'm not using recycled iron on the replacements. 

[Toxophilite]

Sorry I can't be of use with your voltage problem but I have to say that's a damn pretty paint job. Good choice!

[sluckey]

I  did a quick check of B+ and got 700+V

That's theoretically impossible. With 394VAC the max DC you can produce is 394 x 1.414 = 557VDC. I suspect your meter is lying to you or you are doing something wrong.

[Brendan]

I have another meter to try tonight.

[Brendan]

Ok, using a cheapo DMM,  I got B+ of 406, followed by 389 and 347.  The earlier issue with the huge drop of voltage to the preamp section is gone.  The only other change I made was to remove the big ceramic resistor before the can cap.

The 12ax7 is now getting:
1)212v
2)0
3)2.0v
6)217v
7).43
8)1.92v

 It seems to work safely with the 6L6GC. Pins 3 and 4 have 396v and 398v respectively and I'm getting 23w of plate dissipation.  I  have resistors coming so I can dial in the bias  to work for the 6L6GB that came with the reel to reel.  I did a short test but  the only 8 ohm speaker I have is a 12" Cannabis Rex which certainly helped make a loud presentation at 9pm in a city neighborhood.

I'm still not sure what was behind the crazy DMM readings I was getting.   I'm looking forward to building a cabinet for this, so I have some time to adjust and test further.

I really appreciate the help.  Thank you. 

[John]

I'm still not sure what was behind the crazy DMM readings I was getting.

Probably just needs a fresh battery.