Converting an old tube radio into a guitar amp???

[tubegeek]

I'm looking at a Gretch G6151 schematic. There are no voltages listed so there is nothing to compare to, however your voltages look like the amp should function. The one voltage that kind of bothers me is the plate voltage on the first 12AX7. I'd like it a little lower. How does the amp sound if you remove the Tremolo tube? If it gets louder, try changing the 1.5K to 1.2K or 1K.

You're saying that because it appears to be biased pretty "cold," right? I thought so too. I should have more than just a fraction of a mA, right?

I tried something today that didn't really work out but it maybe speaks to your question.

Looking at the Aiken Amps article on designing phase shift oscillators I decided to try bumping up the caps x10 so I could use a dual 100K pot (which is the only dual I have on hand) and see if that was more effective in getting a wider spread between low and high trem frequency. Well, those resistor values seem to be loading down the oscillator such that it doesn't really drive the audio stage very well. The oscillator is definitely starting, I can hear trem, but it's not doing much. As a result, yes, louder audio. But weak trem.

So this version has .33, .22, .22 caps (reading away from the plate) and 22K feeding 100K variable, 22K feeding 100K variable, and 100K shunts. That's a load of, best case with the pots all the way up, 100K||122K||122k, right? Which is like 30K or so. The caps add some impedance to that but then the pots aren't always all the way up, either. So a pretty bad load.

I tried shorting the 2M2, and also both the 2M2 and the 1M feeding the grid of the cathode follower, to try and get some more signal into the cathode follower, but that didn't work much better. Losing too much of the oscillator before that grid network, in the phase shift sections, I think?

I will try bypassing the cathode of the oscillator and see if that helps any with the intensity. Also, as you mentioned, I will try pulling that tube out entirely and see what happens. But I don't dare do that until I bring down the heater voltage to something saner though, it's like 7V now.

One calculation you didn't do was the plate dissipation of the 6V6. It comes in about 16 Watts. A little on the high side. I would increase the 6V6 cathode resistor to 390 ohms and bypass it with 50uF or more.

Thanks for doing that - (Vp - Vk) * Ip, right? I get 16.45 W, hmm, 14 on the data sheet, huh? OK, that needs attention too.

And if you noticed in the calc, I am only using 300 in the cathode so I went the wrong way there - the schematic calls for 330 - so that's at least part of the reason. I don't have anything between 300 and 500 or so in any decent wattage on hand but I can put 68 or something like that in series with my 300, I have that value in a reasonable wattage rating here.

Adding a bypass (10uF) to the cathode of the second 12AX7 stage will give a small increase in gain also.

OK, worth trying for sure.

It's my understanding that the air gap in a single ended transformer doesn't have to be that much. Maybe only the thickness of a couple of sheets of paper. Did you get a chance to try the other Hammond OT?

That's what I understand too - I have only one Post-It note in there now. Did not swap transformers yet.

Thanks for your suggestions, helpful stuff and I will continue to post my progress.

Here is what I posted to the Joenet about this thing BTW, it has some entertainment value if you read to the end:

I got the little guitar amp project up and running and I was not completely satisfied with it at first, so I thought I give a little report on the process I've been through.

Schematic
here, for those who are playing along at home.

At first , this was kinda weak and clean-sounding, which was totally not the point. Fortunately, my conversion omits the two dial lamps and several tubes from the original, giving me about 7V on the filaments. This was helpful because I blew up the weak 12AX7 in the input position that was causing part of the problem. I'll burn off some of the excess filament voltage with a lamp and some load resistors before I kill any more tubes, I promise! A different 12AX7 gave better results.

The OPT is a "Universal" P-P type which I restacked and gapped, changing the interleaving from 1x1 to 5,11,11,5. Not quite butt-gapped, but almost. The first time I tried it, I used secondary taps 1 & 6 and that was too much stepdown and too clean. I moved over to 1 & 4 and the results are better, it loosened up the sound without making it flabby.

Another adventure: I kept letting some magic smoke out at first. Crappy original socket on the 6V6 had an invisible arc-over between plate and screen grid, found it and swapped sockets with a decent ceramic socket, no more smoke.

Tremolo worked really well right from the get go, but I wanted to slow it down a little, so added some extra caps in parallel with what was there already. The two .01's in the phase shift sections of the oscillator are now .02, I may go even higher if it doesn't keep it from starting. I may put in a switch for Fast Range/Slow Range too, to keep all the original sounds available. The trem is VERY good-sounding, I like it a lot.

At this point I wanted to visit my pal who deals in vintage gear - I wanted to hear a REAL player using a REAL guitar on it, and perhaps a decent speaker too.

We hooked up to a 2 x 15 Bassman cabinet which sounded lovely although the amp could not overdrive those of course. One of the crew plugged in a Les Paul Jr, and had a great time playing tremolo'ed riffs with it. We also tried it with the shop's latest prize, a fully restored '72 Mellotron, which of course was drippy and trippy as all get-out. That was pretty cool indeed, and the mechanism and conception of the Mellotron is absolutely fascinating.

Mellotrons appear to use 3/8" audio tape, recorded with 3 tracks. The tape heads appear to be 1/2" though. Is there any other machine that uses 3/8" format? Where the hell did it come from? I doubt you could even get the tape to make the "samples" today. Have any Joes ever seen that format before?

Then Jack White walked in with a few of his pals to check up on a repair and to ask some questions about an unusual guitar he brought with him, an old small oak-bodied acoustic. Best guess was Lyon & Healy-manufactured, circa 1890's. Very sweet-toned guitar. Yes, he was wearing the hat. I think maybe he always wears the hat.

As you can probably tell, that shop is a fun place to hang out!

-j

[loudthud]

You're saying that because it appears to be biased pretty "cold," right? I thought so too. I should have more than just a fraction of a mA, right?

I drew a load line on the plate curves. Because of the 270K plate resistor, the load line is lower than the normal 100K. The load line crosses the 0V bias line about 50V so for headroom, the idle plate voltage should be roughly half way between 292V and 50V or about 170V. The gain should be a little higher there too.

For the trem oscillator I would try a 1meg pot, or possibly a dual. Just keep increasing the caps. When bypassing the cathode, use a large cap, 100uF or more.

[tubegeek]

OK, here's today's action:

1. I put a 200 ohm resistor across the heater circuit, the heaters read 6.2VAC now, so that's just fine.

2. I tried your suggestion to check what happens with the tremolo tube pulled; the result is only slightly more gain (by listening) than with the trem turned off.

(I moved the trem on/off switch from the trem SPEED control to the trem INTENSITY control, by the way, it made more sense to me. The donor AM/FM radio even had the 500K pot-with-switch I needed - I really got lucky with that trash-pick from my friend!)

I think the gain is OK now - I think the biggest problem I had originally was the weak 12AX7 in the first position which I guess got killed by overvoltage on the heaters anyway.

3. I found a dual 1M pot in the basement! So I put the trem circuit back more like the way I had it a couple of days ago: the three phase shift sections are now like this, starting closest to the plate:

.033, .022, .022 caps;

220K feeding 1M variable, 220K feeding 1M variable, 1M5 resistor shunts.

Well, hey now! Slow, nice wide adjustment range, plenty of depth. Very, very nice.

Of course the dual pot I had was audio taper so I installed it backwards as a "SLOTH" control (fast on the left, slow on the right.) This turned out to be interesting - the log taper really does make the control's action seem much more "right." I didn't expect it to make much of a difference but I found I do like it better this way.

Based on what I read in the Aiken article I linked above, I think I'm going to leave the bypass cap on the trem oscillator off. More gain will just add distortion and I don't need more swing anymore - it was a loading problem, not a gain problem. I will try bypassing the second audio triode at some point, see if it'll scream a little more.

4. Changed the cathode resistor on the 6V6 - I used 200 + 125 in series.

6V6 (Power):
Plate 347
Screen 327
Cathode 16.5
Grid 0
DCR of OPT primary 491

Calculations:
Ik= 16.5/325 = 50.8 mA
Ip = (370-347)/491 = 46.8 mA
Is = Ik - Ip = 3.9 mA

Wp = (347-16.5)*.0468 = 15.47 Watts

So the plate dissipation is a little bit better. I'll up the resistance another 30 - 50 ohms and see how it looks.

[Badacatt]

So I got to work at the weekend. First up, three prong cord and fuse. Second, remove the valves from the tuner section. Third, connect a jack to the volume control and hence grid of the triode (6BD7). Fourth, plug in guitar and play. Next time I do this it will take about 15 minutes. I had to understand what was what on this first. I get a decent bedroom volume from this, albeit a bit clean. However, to convert back to original state of a working radio all I have to do is remove the jack and pop the valves back in. As if. Dead easy and dead cheap. Cost me 15AUD for the radio and a buck for the jack. I love building amps from junk. You should give it a try.

Hi Paggerman,

do you have any notes or videos on your modification? You said next time it would take you 15 minutes to modify, and I’m new to this all but I would love a little project like this…

i went out and bought a whole bunch of old tube radios and apparently I’ve made a terrible decision as these all are t the idea for what makes an easy amp mod.

curious what size fuse and type did you use on yours?

any info would be of huge assistance to me as I’m not knowledgeable to this area and I would prefer to live.

thanks again,

joshua

[Chuck H]

Hi Paggerman,

do you have any notes or videos on your modification? You said next time it would take you 15 minutes to modify, and I’m new to this all but I would love a little project like this…

i went out and bought a whole bunch of old tube radios and apparently I’ve made a terrible decision as these all are t the idea for what makes an easy amp mod.

curious what size fuse and type did you use on yours?

any info would be of huge assistance to me as I’m not knowledgeable to this area and I would prefer to live.

thanks again,

joshua

What worked for Paggerman won't necessarily work for you. Not unless you have the same circuit to start with. There are many design particulars that may need to be managed. Even one circuit type called "widow maker" because it can shock or electrocute you. And THESE WERE COMMON! So we really need as many specifics as possible. There's no cut and paste method. The safety end of this isn't hard. The thing to know is that large capacitors in the power supply can hold a charge. Discharging these safely and properly (search here and you'll find good safety info) and seeing to it that the unit is unplugged will keep you safe. It may come up that we need readings from inside a live (turned on) chassis but we'll cross that bridge when it comes up.

Try to get any info you can about the units you purchased. Diagrams, schematics, etc. Short of that you could at least list the brand/model and the tubes in use for each. Start a new thread with this information, an indication of your intentions and mention again that you are a novice. Please indicate the tools you have on hand now (soldering station, DMM, etc.).

[The Dude]

^^^^^^That! Agree with Chuck. I'll add to his "There's no cut and paste method". Since each radio is different (unless you bought all the same model radios), it would be best to start separate threads one at a time for each different radio so as not to confuse circuits. And, as he said, provide as much info as possible on each radio (schematic if possible, tube compliment, pictures, etc.).

[sparkies]

^^^^^^That! Agree with Chuck. I'll add to his "There's no cut and paste method". Since each radio is different (unless you bought all the same model radios), it would be best to start separate threads one at a time for each different radio so as not to confuse circuits. And, as he said, provide as much info as possible on each radio (schematic if possible, tube compliment, pictures, etc.).

Very correct. As not all are made the same and some like zenith radios had high voltage on the chassis that we referred to as a 'hot chassis' So there is use cases.

But there is a few things that should be established in a portable radio, since you will be installing a guitar jack. One would be changing the cord to a 3 prong cord and bonding the earth ground (green wire) to the chassis, and grounding one side of the speaker to the DC ground in case of failure, it would pop the fuse. Also, you would tie dc ground to the chassis.