Converting an old radio - Tremolo help?

[johnnyreece]

So, after a year of hemming-and-hawing (I'm really good at that), I finally got off my duff. I reimagined this with noval tubes. It's mostly a Vox AC4tv, but I've used Sluckey's tremolo circuit, SS rectification, upgraded filtering, and a 5u8 (6u8 in the layout). I had a few 5u8 on hand that tested well (and had ditched the rectifier), so figured I'd use what I had. Can always rewire it later. The 5u8 has a pentode/triode combo. I'm hoping that extra triode gives a bit more oompf than the original. I've got a turret board made up and mounted, and I'm starting to get the rest of the hardware installed. I'll post pics when it's a little further along. In the meantime, here's my updated layout. In hindsight, I'd have done things a little differently, but oh well...live and learn! One thing I noticed is that I erred when doing my power supply...will probably ditch having a C node, and have B become A, with C becoming B. Will update the schematic/layout when I have a "final" version.

[johnnyreece]

Okay, so I lied...found a way to make the C node work. Layout updated, and here's my cobbled schematic.

[R.G.]

i did a similar conversion back in the mid 90s. It's a great approach. I especially like the style with a single ended EL84. Many, many of this kind of thing were popular in the late 50s and early to id 60s, and they are indeed frequently free or only a nominal price.

My amp produced a robust 4W, but had the sweetest slide into overload and distortion I ever heard. I loved it.

Even if you're a club/gigging musician, one of these feeding the PA or another power amp would be a great alternative.

[xtian]

Based on your schematic, I think this is going to have very wooly distortion, possibly not very articulate because of the full bandwidth. You might need to swap C3 with 0.68uF, and maybe add a treble peaker or bright cap. But that's certainly stuff you can mod after you get it up and running!

[johnnyreece]

Thanks for taking a look. I lifted the preamp from an amp design called the "Savage Croaker", so you could very well be right about it not being tame. This is actually for my wife, who thinks my 2w Express is too loud. I may be pretty off-base with the character, but there's a lot of old radios out there if I don't get this one right...

[Phil_S]

Congratulations on getting this done. I love this kind of project and I think the results can be very satisfying in a number of ways.

[johnnyreece]

So, had to spend my afternoon at work watching training videos we watch every year, so...perfect time to update layout/schematics!

[johnnyreece]

So, finally an update on this. It *does* work, in that it puts out guitar sounds. It has some issues, though. Right off the bat, I noticed I mistakenly installed a .047, rather than a .0047 cap for C1. Aside from that, here are my first impressions:

-It is *extremely* bright.
-The distortion is (as xtian supposed) quite wooly. I will see about swapping C3 for a .68uF (or dropping it completely).
-The tremolo doesn't work. If engaged, I get a loud thump at minimum intensity, which decreases in pitch and speed when the intensity knob is turned up. Speed control seems to have no effect. Swapping in a 12AY7 seems to cure this, but then there is minimal (VERY minimal; maybe imaginary) tremolo. I'll check this over with voltages when I get a moment. Is it possible I need to connect it to the triode cathode, rather than the pentode?

I've attached a picture from when it was *mostly* done. All that was left was the 1 meg on the input, heating my shrink tube, and the power cord. I think it has good bones; just need to iron out the details. Side note, I know I don't have a shorting jack on the input. I'll actually have a shorting jack mounted on the radio that will connect to the jack on the chassis with a jumper. I know it's a little convoluted, but it should work.

Any advice is welcome!

[johnnyreece]

I got some time to take some measurements. Note - this is without the tremolo on. Other than my filaments being WAAAAY too high, does anything else look suspect? I'll try to take a look at it myself, but it's time for JEOPARDY!

[Phil_S]

I'd say you can tame the filament voltage with a couple of sandblock wirewound resistors. Put one in line with one leg of each filament winding.

It's been too long since I calculated what to do. I'm not sure if I'm remembering correctly, and the math I use is imperfect anyway because I ignore the actual voltage. You have 5V and 6.3V supplies.

The 5U8 uses 0.6A. So, 0.6 * 5.0 = 3W. Use at least a 5W resistor. You need to drop about 0.9V @ .06A. V=I*R; R=V/I ; R=0.9/0.06 =15Ω.

The EL84 + 12AX7 uses 0.76 + 0.3 = 1.06 @ 6.3V = ~3.5W. Use at least a 5W resistor. You need to drop about 1.13V @ 1.06A; Ohms law says 3.3Ω. Not sure, but I think this may be a standard value.

These resistors will get hot. I'd consider 10W rated. Make sure they have enough free air around them.

Just a sanity check here. If we assume this is a 110V transformer, it means your line supply is about 125VAC. That's not unheard of these days. I see that at my house.

For the tremolo to trigger, you will need a strong 12AX7. Use this tube for it's high Ra = 62,500. The other 12A?7 tubes have Ra that is too low. If you have a known good tube in the amp, then it is some other problem. Merlin has a good article on tremolo. http://www.valvewizard.co.uk/trem1.html He says you can also use a 12AT7...I guess he knows.

[johnnyreece]

Thanks for taking a look! Yeah, I forgot to include my wall voltage, which is 122.5-ish. I'm sure part of the issue is that I'm under-utilizing the filament taps. For example, the 5v tap is for a rectifier, which I believe is rated at 2A, and I'm only using .6A. After looking at that link, I'm guessing maybe my plate voltage is too low for the tremolo. I may need to reevaluate values. The tube is a known good one. I'll keep at it.

[Phil_S]

In a sense, the in line resistors are helping to bring the load to what was intended, but your observed voltage is so high that lack of load can't be the only explanation. At 122.5V, line voltage is 106.5% above 115V. (If your readings are based on 122.5V input, your PT is probably designed for 115V, not 110V.) 5.89V is 117% of 5.0V and 7.14V is 113.3% of 6.3V. Most of that overage is assumed to be attributed to line voltage.

If you can get to 5.2V and 6.5V, that is more than close enough. It's going to fluctuate with wall voltage, which varies throughout the day. Go ahead and check the wall voltage several times over 2 or 3 days and you'll be shocked (figuratively; pun intended.) I've done this before with old iron. It works OK. I should note, I don't gig, so I'm not really sure what happens if you run the amp hard for 2 or 3 hours. I reason that resistors overrated for watts should work well. In this case, bigger is better.

If you work on the tremolo cap values, you'll get it working OK. You might look at Fender VibroChamp to see what that amp used. Maybe it will help. I don't know if it is the same type of tremolo.

[johnnyreece]

I think it's based on the Vibro Champ. It's a design attributed to Sluckey. I was hoping he might pop in and give me the easy answer as to what's wrong, but I suppose I haven't given much information. I have a clip of the noise, but need to get it hosted somewhere and in the right format. Even with the volume all the way down, I get a loud (sorry for the description) slappy fart sound that gets deeper when I turn up the intensity knob. Speed seems to have no effect. If I can't get this sorted, I'll likely go for a more "stock" setup, like the Vox AC-4 (which is what the power section is based on, anyway).

[johnnyreece]

So, my understanding of the tremolo circuit is getting a little better. I have realized, I have it backwards...it makes the awful noise when it's being grounded (the trem is off). Is it possible that a ground issue can cause something like that?

[johnnyreece]

Okay, so after some tweaking last night, I have tremolo! I think I had a grounding issue. I removed a duplicate ground, and moved the plates to the A node. I need to tweak some values, as I get a reasonable range on the slow speed (would like it slower, but we'll see), but the fast speed is just too fast throughout the travel of the pot.

Side note, the slide into distortion isn't pretty on this amp, but once you get past that, you can get a really brittle, attack-y distortion. I didn't really expect it, but I don't hate it! I wish the in-between was better. It's a bit farty before it gets there, though. I think first thing's first, I'll remove the bypass cap on V1. Will take it from there. When I get done, I'll update the schematic/layout to the final version, in case anyone else wants to try this out. Still need to get the filaments sorted out. Will get some resistors ordered.