Finally Building JTM45: Stupid Questions As They Arise

[The New Steve H]

If I hated math, I would never have gotten a physics degree! I probably said I hated engineers for one of their weird habits, like putting 5 variables in a 2-axis graph. Or maybe it was just the way they dress.

Got myself a Radio Shack meter. Very exciting.

[Reeltarded]

I just got a debug message on a PM myself.

Steve, here is the biasing portion of the Metro 1987 instructions. Not meaning to lift a likely copywrited material.. it's freely available, as is the JTM45 instructions on George's Wiki.

The only difference will be your KT66 vs EL34, your voltages, and where the quiescence be.. at.

(Prepositional ending to a great post.. plus I used a big word)

[The New Steve H]

Thanks. I was going to try my socket idea, since it's easier than turning the amp over while it's running, but it looks like I don't have any spare sockets. I can't find a 1% resistor locally, so 5% will have to do.

I can see that my friend is going to have to bring this thing back to me so I can do it right.

[Reeltarded]

Ummm 5% is probably close, but not close enough. Measure til you find one dead on for each tube. The tubes might be a couple or more mV from each other, but those resistors need to be perfect so we know.

[The New Steve H]

This amp has pins 8 and 1 jumped on the power tubes. I guess I should remove the jumpers and insert 1-ohm resistors instead of wrecking the nice ground wires I put between 8 and ground.

[Reeltarded]

Go follow the instructions of the 50 watter @ Metro. The JTM has a previous version of installing that hasn't been updated, and the pictures are very clear.

[overtone]

Thanks for the help.

I had an idea for a gadget. You mount a socket on some kind of base, and you solder wires to the bottom, leading to pins which fit in the amp socket. You rig it up so one wire goes to your ammeter and then back to the amp, so you can measure the current while biasing. You put the tube in the new socket, and you plug everything in to the old socket and the ammeter.

Bad idea?

along time ago I made one of those gadgets and called it the Octopus. It had four octals linked by bundles of trailing wires to four octal plugs. A fat load of good it was. The octopus and I had some gripping horror B-Movie battles. Eventually I won, chopped it up and have been living happily ever after with the simple 1 Ohm resistors.
Best, tony

[The New Steve H]

Radio Shack has no 1-ohm resistors, unless you count giant ceramic ones. Best I can do is 10-ohm. I have no idea how much this will affect the current, because I don't know the resistance of the crap downstream from the cathode. I think it's a lost cause. One ohm is negligible. Whether ten ohms would be negligible in this position, I can't say.

This is going to be a pain. I'm going to have to disconnect the cathode leads and insert alligator-clamp jumpers while I check the current. I'll have to prop the amp up on its side in order to do this.

[The New Steve H]

Okay, I just did V5. I lifted the ground on the cathode and put the ammeter between the cathode and the chassis. The most current I could get was about 33 mA. I used an online calculator, and I got about 42 as the 70% figure. The B+ is 420V.

[Reeltarded]

Measure them. If they are 10R on the nose, perfect.

Now you are loking for readings exactly one order of magnitude from 1R. The only difference is a zero.

Ohm's Law, not Ohm's really strong advice!

[The New Steve H]

Oh, boy. I'm afraid my friend bought the wrong PT. I never checked it, since he researched it himself. This thing is a Classictone 40-19033. The spec voltage between the red taps is 690V. I'm getting 420 for the B+, but I am reading that this is pretty low. I'm having a hard time finding a suggested B+ for these tubes.

The fixed resistor in the bias circuit is 82 ohms. Maybe I should change it?

[M Fowler]

That's the transformer I used it should read 40-18033 at 690vct not 40-19033 though.

Instructions from Triode:

Set your meter to the Highest DC volts setting (600V or higher), and put
the common (ground) lead on one of the chassis mounting nuts or
alligator clip it to the chassis. Touch the positive lead to Pin 3 (plate) of
one of the KT-66 power tube sockets. This is your Plate Voltage. Write it
down.
*BE VERY CAREFULL WITH THIS STEP (HIGH VOLTAGES PRESENT)*

Most KT-66's are 27 watts maximum dissipation and you want to set the bias to 70% max dissipation. (70% = 18.9 watts) Divide your desired wattage by the plate voltage that you wrote down from step 2. (example: 18.9/450 = .042 or 42mV) This is your desired bias voltage (probably between 35 and 45 mV).

Set your meter to lowest DC Volts setting (200mV or less) and touch the
common (ground) lead to the grounded side of the 1 ohm resistor coming
off of Pin 1 and the positive lead to the Pin 1 side. Adjust the bias
potentiometer until the bias voltage you calculated from from Step 3 is
achieved.

Plate voltage will change a little when adjusting the bias, so you may need
to take another reading from Pin 3 and adjust your equation accordingly
and repeat steps 3 & 4.

[Reeltarded]

Why would they say to put the negative probe anywhere near the positive probe, inside an amp with 450vac just half a probe away?

I would clip the - to the chassis ear in this case.

[The New Steve H]

I'm measuring the current directly because I have no 1-ohm resistors. My friend is bringing the cab tomorrow, so I told him to get some resistors. The number is 18033; my typing is not always perfect.

The amp sounds great, but something is going on with the normal channel. It sounds muffled. I'm going to have to poke around in there.

[The New Steve H]

This amp has two sets of inputs. Two normal and two bright. Is it imperative that they have separate grounds? I ran a wire from the normal jacks to the bright jacks and then to ground. The normal channel seems muffled. I can't figure out what else would cause this. I don't see why it would matter, but the sound problem is there, and I've checked all the connections on the board and pots.

Also, give me some help on wire dress. The OT leads are loose because I wanted to be 110% sure I had them connected to the right pins before I shortened them up. Now I'm taking up the slack.

I know you're supposed to twist these together, but obviously, you can't do that all the way to the sockets, because the sockets are a couple of inches apart. Does it matter if the twisted stuff ends midway between sockets, or can one lead be shorter than the other? An asymmetrical arrangement would be more convenient.

Posting a photo. The red and black wires are the ones I'm talking about.