5G9 Tremolux PTP Build

[Phil_S]

Here it is, warts and all. I haven't yet verified the build to the schematic, something I like to do before power up. I think it will be at least a couple of days until I can get to that point. But I thought I'd share. The chassis came from an organ, so socket placement was predetermined. Working around that...let's say it was fun. PT and OT came with the chassis. The choke came from the bone pile -- orange wires coming up under the yellow tremolo board between f/c1 and f/c2. Please be gentle! OK, you really don't need to pull any punches. Thanks for looking.

[tubeswell]

Hi Phil, not quite the layout I was expecting but hey, have you fired it up yet?

Pete

[Alexo]

The proof of the pudding is in the eating!

Looks like a fine PTP to me... bet that IEC socket was fun to cut out from that old plated steel, or whatever super-metal they used on some of those old chassiseses.

[Structo]

Cool!

What did that start out in life as? TV?

[Phil_S]

Hi guys...
Pete: So, how about the layout! Not fired it up yet. Still need to verify to the schematic one last time. I'm a careful type. In the final soldering, I found I had messed up the PI filter in the bias circuit, which is why you see a free hanging 8u next to the PT. I did what I could with it.

Structo: It was an organ with three channels, two SE EL84 (Mullards!) and PP 6V6 (Tung Sol); tubes came with it.

Alexo: Yes, will find out, probably not for a few days. I cut the IEC and a few other cuts with a Dremel loaded with a cut off wheel. After a while, the wheel gets smaller. I used an old one to make small cuts and then finished it with files. It actually went quite well. Yes, super metal! I had a good time making the IEC and I shaped it to the connector instead of making a rectangle!

Trying to keep the smoke in!

Thanks.
Phil

[Phil_S]

Tom: I see your question...the bit of metal I used to mount the pots, that came from an old Heathkit TV chassis. This chassis was sitting there on the curb one Saturday maybe 7-8 years ago. I drove by it a number of times that day, as it was near my house. On my final trip of the day, I brought my cutters and extracted the chassis from whatever cab it was in. Later, I cut it up and build an amp in it with the PT that was attached. Another story, another time. Anyhow, this was a scrap left over and sitting in the basement waiting for a place to fill.

I caught a break tonight and got some unexpected bench time. I verified parts and wiring. That amp wouldn't have made even a peep. The coupling cap from the 1st stage to the volume pot was missing. I managed to shoehorn it in there.

While I was building, I completely missed the correct value for the bias supply dropping resistor. The schematic pulls voltage from a bias tap, but my PT has none. I've got to do this Marshall style. I upped it from 82K to 100K figuring this PT is going to give higher voltage than the schematic. That was all wrong. I probably need 180K-220K to get a proper drop for the bias supply. As long as I've built it out that way, I'm going to fire it up and see how bad it is. I've got 47K+25K pot on the ground side, so I may be able to compensate. How many times am I going to change that resistor? Once if I can manage.

BTW, never trust your memory for details. I want to say that PT is 325-0-325, but I should have written it down because I really can't remember and now that I have it in circuit, I've got to power it up properly. I'll be using the dim bulb method. And I've got my new old Fluke for voltage checking! Nice meter. Now I know what you guys are talking about.

Well, that's it for the verification. It pays to do this, and is probably doubly important on a PTP build where there is no layout diagram.

[diagrammatiks]

fire it up!

[tubeswell]

While I was building, I completely missed the correct value for the bias supply dropping resistor. The schematic pulls voltage from a bias tap, but my PT has none. I've got to do this Marshall style. I upped it from 82K to 100K figuring this PT is going to give higher voltage than the schematic. That was all wrong. I probably need 180K-220K to get a proper drop for the bias supply. As long as I've built it out that way, I'm going to fire it up and see how bad it is. I've got 47K+25K pot on the ground side, so I may be able to compensate. How many times am I going to change that resistor? Once if I can manage.

BTW, never trust your memory for details. I want to say that PT is 325-0-325, but I should have written it down because I really can't remember and now that I have it in circuit, I've got to power it up properly..

The 5G9 schematic suggests 300-0-300 for the PT and this works well with a 5U4GB to give about 370VDC. I built one in 2008 using 300-0-300@150mA, 6.3VAC@3A, 5V@3A with a 50VAC bias winding, and it hit the 'mark' pretty much perfectly. Yep - do an adjustable bias

[Phil_S]

I figure, this is the iron it came with. The rectifier in use was a 5U4GB. It was using 6V6's. I figure it will be OK. I'm curious to see, with the unknown choke, what the screen voltage will be.

There is no way the bias supply will be right. I've got to put a 180K or 220K between the HT secondary and the diode or it will probably blow up! Right now, I've just got the diode sitting directly on that 300+VAC line. One more fix up needed before I can power it up.

[Phil_S]

I got to thinking about this. I powered up, no tubes, just to check the PT. Unloaded voltage is 325-0-325. If the PT is 117v rated, then today's voltage is about 3.5% higher. Under load, it is likely to drop about 4%. I think that PT was rated either 300-0-300 or 305-0-305. I am going to guess plate voltage is going to be above 370, and certainly no higher than 380-385-ish.

[tubeswell]

Depends one the current draw rating of the secondary windings. With 325-0-325 and a 5U4G, and 2 x 6V6s, you may get around 420 B+, so you may want to use JJ6V6S

[Phil_S]

I think tomorrow I'll get the real answer. Today, here's the math I'm doing to estimate, several ways.

Assume it's 325-0-325,
A) 325 * 1.4 = 455; 5U4GB should drop 50V. Result 405 is B+, maybe 400 plate voltage after the OT takes a little.
B) 325 *1.2 (common rule of thumb) = 390. This isn't far from 400-405-ish.

Assume loaded voltage is 312-0-312, a 4% drop under load (drop under load is a rule of thumb/guess).
C) (312 * 1.4)-50 = 388
D) 312 * 1.2 = 374

So, I'm guessing B+ will be in the 380-390 range. I'll be surprised to see 420. I guess we'll find out soon whether any of this sort of estimation holds any water. Right now, I suppose this is simple banter.

[Phil_S]

I'm self employed. I played hookey this morning even though I should be working and got some voltage readings with tubes. I didn't run a guitar or other signal through it, as I think it still needs a bit of work. I'd appreciate some input. The good news is that nothing burnt and the pop test indicates the signal chain is working. Pulling the shorted plug from the tremolo jack turns off the tremolo, too!

Schematic voltages inside [xxx]; A=plate or anode, G=grid, K=cathode, S=screen

V1, 12AY7, omitted the "mic" channel, and paralleled the sections, 56K plate load, 470R/25u on the cathode.
Plates 120 [175]
Cathodes 1.5 [2.5]

V2, 12AX7 (LTPI) Japanese Mullard? Came with the chassis.
A1, 195 [210]
G1, 15.3
A2, 187 [200]
G2, 16.3
K, 23.5 [27] common to both sections

V3, 12AX7, Tremolo
A1, 250 [270]
K1, 1.6 [1.7]
A2, 358 from screen supply [368]
K2, 252 [260]

V4 & V5 are Tung Sol 6V6GT's that came with the chassis.

V4:
A, 343 [not given, but using OT that came with chassis; see Rectifier and B+ readings below]
S, 358 [368; see below B+], screen current 1.6mA
G, -26.5 minimum setting on bias pot [27.5]
Bias over cathode 1R, 46.4mA

V5:
A, 346
S, 358 [368; see below B+], screen current 1.4mA
G, -26.5 [27.5]
Bias over cathode 1R, 40.6mA

V6: 5U4GB, came with the chassis
Secondary PT supply 645VAC
B+1 358 [370]
B+2 358 [368]
B+3 304 [310]

5V filament, 5.1VAC
6.3V filament 6.55VAC

So, here are some questions/observations/comments. Your reactions and advice could be very helpful.

I'm not happy with screens running above the plates by 12V. I am using a mystery choke. I'm not wanting to face the music here? Testing suggested it is 3H, but I'm not sure. It is fairly heavy, so I took a chance. This choke isn't doing the job, is it? I need to try another choke?

I've got three other chokes:
-mystery choke that testing gave a guesstimate of 30H, it is 556 ohms. This is the one size-wise that will be best fit. Since this is the screen choke, I don't think I've got to be too concerned about mA rating?
-choke for a Vox AC30, 20H, 100mA. This one is large, with a 4 bolt mounting pattern and will be difficult to fit, but not impossible.
-Stancor 4H 250mA big and heavy, too hard to work onto this build.

I was a little surprised by how much drop there was from 645VAC to B+1 at 358VDC. <shrug> No big deal, but wow, not what I expected, see my guesses in the other post.

I wasn't happy with the range of voltage on the bias supply. It goes from about -36 to -26.5. I think I should reduce the value of the first dropping resistor in the PI filter from 100K to 82K and I'll be good to go, or else swap 150K for the 180K that connects to the HT secondary. I was betting on more drop that I got there. Unloaded it is about 325VAC on one leg. Under load is only dropped to 322VAC.

With respect to the HT secondary drop under load to 322VAC per side, is there anything I should make of the lower than expected B+1 voltage? Where I'm weak on understanding is the effect the screen choke might have on this.

Regarding V1, should I be looking to increase plate voltage? If so, what's best to do? 68K plate load? 680R cathode? Some other values? both?

Thanks for looking in on this one.

Phil

[martin manning]

Nevermind... Read a little more closely.

[Phil_S]

Martin: You are a tease
So I took your advice "read" (say "reed" not "red") and I did that. I faced the music with the choke and put in the one I guess is 30H. Had to drill one mounting hole....eeeks! That gave me a reasonable drop between plate and screen of around 6V. Of course, everything down the line is down 6V or so, as well. I really should bias a little colder. I've got it at 38/43mA.

Now, I'm wondering mostly about V1, whether I should do something to increase plate voltage or what.

The amp has a little 120Hz hum on the bench that I'm hoping won't be so noticeable after I button it up. If I really don't like it, I'll bring out the scope. That will be a learning experience for me.