5E7 Modifications

[JohnLochner]

Hello all,
I have recently completed a build of a 5E7 circuit, and I made a modification along the way.

Before I go on, let me say that I am in general not a fan of Master Volume controls in tube amps...I have a Budda Twinmaster and a homebuilt "Double-Champ", neither of which have Master Vol controls. However, I am in need of an amp that I can take to super-small gigs and still achieve some amount of crunch, so I decided to build a 5E7...with a Master Volume.

What I did was to eliminate the second channel of the 5E7 circuit, and use the second existing "Volume" pot as the Master Volume.

Then I got to thinking- I now have a triode section (in the first 12AY7) that is no longer used.

So, I "jumped" the triode sections of the first tube EXACTLY like a 5F2, with a Volume control/.022 cap between the first plate and the second grid.

The one thing I did not change was the cathode Resistor/Cap arrangement on this tube. I still have pins 3 and 8 shorted together, then going to a resistor/cap, then to ground. This is how the original 5E7 was wired.

My problem is this: the amp works ok at low preamp levels. Above halfway on the preamp control, the amp goes into self-oscillation, and howls uncontrollably.

The howling is not dependent on the Master Volume control. Less Master Vol equals the same howl, only quieter.

The tone controls affect the howl. Turn the treble up, the howl gets "more treble-ly".

I reversed the connections to the primary side of the output transformer, as was suggested to me in the case of a howling amp, but this did not change anything.

I believe I have isolated the howl to the preamp section. However, I'm not sure what is causing it. My only suspect is the shared-cathode arrangement of the first tube, but I am curious if anyone else can shed any light on this.

And as always, I appreciate any information you may have!

Regards,
John

[JohnLochner]

Re: 5E7 Modification

My apologies, the first 12AY7 is wired up exactly like a 5F1 (the Champ), not the 5F2 (Princeton).

[JohnLochner]

FYI, Tests conducted

FYI, in case you are interested...

I went ahead and removed the shared cathode resistor arrangement on the modified 5E7. Instead of one 820-ohm resistor hanging off pins 3 and 8, there is now one 1D5K resistor on each of pins 3 and 8. No more sharing cathode resistors.

And, there is no change to the self-oscillation problem.

So, I'm off to check other things. Thanks for looking... just wanted to post my findings.

[Ray Ivers]

John,

If you're using that 10meg feedback resistor from the 5E7 circuit, I would try pulling that out and see what happens.

Ray

[MWJB]

I may be misunderstanding something, but what you say (John) may be interpreted in two ways. You say the 1st triode is configured like 5F1, does this then feed the 2nd half of V1 or feed V2? The phrase "jumped" might suggest that you have tied both plates, grids & triodes together (like a BF reverb driver), but a 5f1 doesn't have this.

Most people would assume that your master volume is situated after the tone controls and before the PI in this sitiuation (like a BF volume control/Marshall 2 input master). Is this what you have done?

If you have simply tacked another stage in front of the existing preamp, with a 1meg pot, you will have massively increased gain. Please clarify.

[JohnLochner]

Re: V1

MWJB,
You're exactly correct, I tacked on another gain stage in the preamp. I'm using both triodes of V1 for preamp gain, then the circuit is stock through the tone stack.

The Master Volume is post-phase inverter, as described in the 5E7 section of this website: http://www.kilback.net/homebrewtweaks/amps/amps.htm

I went ahead and disabled the extra preamp gain stage, and lo and behold, the amp works fine. I think the 5E7 layout is just not up to dealing with extra gain in the preamp, specifically with respect to lead dress/parasitic oscillation problems.

So the amp is running well now, with one exception (and I'm going to post this on the Weber boards and see if anyone else has heard of it).

The overall low end (bass) in the amp is strange. Any audio that has low frequency content is almost delayed- sounds like a capacitor is taking time to charge up, or something like that.

When I play low notes on the guitar, it takes the amp about 1/10th (or less) of a second to actually amplify the note. Also, I have a very percussive style with leads, so my picking transients get lost, and if I play notes fast enough, the actual notes will get lost (because I am holding the note less than the time it is taking the amp to figure out how to amplify the bass frequencies from the transient).

If I turn the bass control all the way down, the amp works perfectly. It just has no low frequencies, but it works normally.

If I turn the bass control all the way up, the amp will sometimes cut out entirely.

My immediate suspect is either the power supply or the capacitor on the bass tone control.

Tubes are as follows:
5U3 rectifier
JJ 6L6, plate voltage = 413VDC, cathode current = 43ma
JJ ECC83 in all preamp/PI positions

Thank you again for reading this far down...
-John

[markbul]

Just a thought: instead of being "delayed", your notes may be cutting out when they exceed a maximum amplitude. So when you dig into a string, the initial attack disappears, only to come back when the signal falls off. That might explain the low notes cutting out - low frequency strings tend to produce bigger amplitude signals.


The cause? I dunno. It almost sounds like an extreme form of sag. Do you have a standard size power transformer for that amp? I'd start by making a list of voltages for the entire circuit to make sure there isn't anything obviously wrong.

[MWJB]

I'd back off the bias to 30-35mA.

You may also want to try a 12AY7 or 5751 in V1, it's possible that with the low voltages you have, the bias of V1 is off just enough to drag slightly when hit with a big signal? The lower gain of these tubes will liven things up a bit, less stodge. Alternatively, try upping the value of the V1 cathode resistor (2.7K or next highest value and halving the size of it's cathode bypass cap, say 10 to 15uf).

If that doesn't cure it you could have a parasitic due to layout/routing problem. The master volume control on this link you supply doesn't appear to be in the optimum place on the control panel, it should be as physically close to the relevant stage in the circuit, certainly not up by the inputs. Disconnect it and see if you still have the problem...in fact do this BEFORE the V1 mods suggested just to rule it out.

[JohnLochner]

Fixed the Problem!

Just wanted to post the results of my fix, in case it helps anyone else here.

I moved the center tap of the primary side of the output transformer before the choke. This is the way a 5F6A is wired.

Turns out I was experiencing tremendous amounts of sag. I like to play tube amps when they are maxed, and apparently this caused grief in the power supply. The mod above fixed the problem completely. I'm now running flat-out, even with a 12AX7 in V1, and the amp holds together nicely. Still has good low end at high volume, though the bass control does not need to be set above "2".

Full credit for this idea goes to Paul, over at www.specialtyguitars.com. A very nice guy, if you have time, check out his website. He dropped me this little tidbit over on the Weber boards.

After the mod above, I went ahead and rebiased to a more appropriate level. At 417 VDC on the plates, bias is now set at 50ma. This change revealed the rest of the tone in the amp...it now spanks very nicely, even with the JJ tubes (they're current production, assuming they're not as good as NOS, though I have no personal basis of comparison in this particular amp). I've used NOS pre and power tubes in my Budda Twinmaster, and the biggest difference there was power tubes and rectifier. Now I use NOS tubes only in the power section of that amp, and leave JJs in the pres.

Also, I went ahead and shellac-ed the tweed, using the 50-50 Zinsser Amber Shellac/Denatured alcohol mixture. I can HIGHLY recommend this to anyone with new tweed. Two coats, and it looks quite authentic.

I had read that you should "prime" the tweed with a coat of clear shellac, then use amber shellac to get the color to be more even (since that first coat soaks into the tweed so quickly). I opted not to do this, because I was looking for a little "nonlinearity" in the color, and can highly recommend to anyone wanting a less-perfect finish. I think it looks better when it's not so even. The results of not priming the tweed were exactly right. Makes the whole thing look a little "older".

Thank you for all your posts and thoughts (and time) here. This board is, as always, a tremendous help!

Best regards,
John

[Bruce / Mission Amps]

I am suspicious it is not the choke location so much but instead, an under rated choke, power transformer or very soft rectifier.
What choke, PT and rectifier is in this amp that would create this much sag?
I read 5U3.
Is the rectifier a 5V4, or a 5U3, which is like a russian GZ30'ish 5Y3GT?

The bigass Heyboer choke I use in my tweed 6L6 amps is rated at 200ma has a DC resistance of around 60-70 ohms.
The Heyboer power tranny I use is rated at 340vac@200ma while under load.

A 200ma draw through the Heyboer choke would only drop the DC down by 12V-14v and I doubt there are many among us who could hear sag in the B+ rail dropping that little.
But a soft rectifier would drop way more and a too small of a choke woudl really hurt too.

[galibier_numero_un]

The bigass Heyboer choke I use in my tweed 6L6 amps is rated at 200ma has a DC resistance of around 60-70 ohms.

Low DCR ... amen, brother Bruce.

Cheers,
Thom

[JohnLochner]

Hi Bruce,
Thanks so much for your message.

The choke and PT are Weber products. Choke has the following spec: 5-8H @ 200ma, 125 ohms.

Noticing that DC resistance is around double of the choke you are describing. Hmm.

There are no quoted specs on the PT on Weber's site.

Before I made the mod I described, I tried three different rectifiers. First, the Copper Cap rectifier that Weber ships with their kits. Second, a Groove Tubes 5U4GB rectifier. Third, a NOS 5U3, which is Russian-built plug-in replacement for a 5U4GB.

All three recti's are good- I tried them all in my Twinmaster, and they all work (though the 5U3 is my preferred choice in that amp).

The amp exhibited the problem with all the rectifiers, with the 5U3 being very slightly worse than the others.

I think the DC resistance of that choke is probably in question here. Also, I personally question the PT, as the plate voltages I get are about 417VDC, with a wall voltage of 122VAC. Seems to me the plate voltage should be a bit higher.

I've heard great things about Heyboer and Mercury Magnetics products. Looks like a choke may be my next thing to focus on.

Thank you again for your reply!

Best,
John