B+ Connection JTM 45

[Ed_Chambley]

Now that I have the amp running nicely thanks to help from the forum.  I have a question on wiring a Hall VVR 3.  Where is the best place in the B+ to break to make a send and return for the VVR?  I am only wanting to dial down the power tubes as I am also installing a master volume.

Also, my bias completely down is giving me -49.  I am assuming that to be a high for el34's.  I am using the mojo layout and the Bias dropping resistor is a 220k.  Anyone have an idea of what value I could change the resistor to.  Plate voltage is 462 so -38 at the pot center of travel would be ideal.  At least that is what I am thinking, but I am probably wrong.

Any suggestions?

[tubeswell]

my bias completely down is giving me -49.  I am assuming that to be a high for el34's.  I am using the mojo layout and the Bias dropping resistor is a 220k.  Anyone have an idea of what value I could change the resistor to.  Plate voltage is 462 so -38 at the pot center of travel would be ideal.  At least that is what I am thinking, but I am probably wrong.

Any suggestions?

To bias the EL34s hotter, you need to reduce the resistance of the tail/range resistor that you are using in conjunction with the bias pot.  On that layout you posted, this is the "47k-82k" resistor going from "E" to ground, (the "E' side of this resistor being attached to the 10k bias pot's 'ground-side' lug). Whatever size you have in there now is too high-a-resistance to allow the bias pot range to effectively decrease the -ve bias voltage any further.  If this resistor is swapped out for a lower value resistor, it will decrease the overall -ve bias voltage range available within the pot dial. I suggest maybe reducing the resistance value by half and seeing where that gets you to begin with. You don't want to go too low though, or the EL34s will red-plate.

On the other hand, if you don't go low enough, you'll still be biased 'too cold' (however, not having heard your amp myself, I don't know whether its running okay or not. Cool biasing doesn't necessarily sound bad, and it makes the tubes last longer).

Ideally you should have the bias voltage set so the the tubes are operating in Class AB1. This mean that the tubes should be idling at somewhere between 60 to 85% of the tubes' recommended maximum plate dissipation (which for an EL34 is about 25W - so you want the tubes idling at somewhere between 15W to 21W). Exactly what -ve bias voltage this ends up being at the grid is somewhat academic, because whatever the -ve bias voltage is, it is the plate dissipation that is critical. Having said all that, it is likely to need to be somewhere between about -10V to -20V or thereabouts.

[Ed_Chambley]

The bias dropping resistor was keeping the bias at -49vdc with the pot turned all the way down.  I thought the same thing, but changing it to a 27k did nothing.  At first I thought raising the value 220k dropping resistor.  Added a 270k and the lowest achieved was -61.  Put a 67k in series with a 100K and bingo, I got -38 at about center travel of the pot.   I measured the pot alone and got 4.88.  This with the 47K still connected to the 1st lug on the pot.  Could it be the 15k resistor which limits the pot and the 47K is the negative feedback?

Anyhow, if I've got something wrong, please let me know.  I have repaired many amps, but built only a few and this is my first marshall build and the bias circuit I am used to has a 50v tap as my favorite amps are the AB763 fenders, but I wanted a British sound.  Got this amp feeding a 2, 12's with weber blue dog and silver bell.

In any event, the plates measure 462, the bias is set at -38, NOC Mullards fx2's and 3 RCA Black 12Ax7's.  It sounds wonderful as I am amazed.  Installed the VVR and it sounds like S**t.  Touch sensitivity was gone and it sounded like an Ampeg VL 50 watter.  I pulled that thing back out.  Whatever the Hall Amplification VVR is for, it certainly ain't for a JTM 45.  I am going to use the chassis hole from the VVR for a master volume.  I had a MV wired as the directions from Hall stated it needed one if you are to control the power tubes only.  I am kind of feeling the VVR is not for me,  I read so much info on the VVR like it was the greatest thing ever.

Sorry for the loooog post, I am just really excited.  Thanks to all on the forum who have helped by answering my many questions.
 

[tubeswell]

The bias dropping resistor was keeping the bias at -49vdc with the pot turned all the way down.  I thought the same thing, but changing it to a 27k did nothing.  At first I thought raising the value 220k dropping resistor.  Added a 270k and the lowest achieved was -61.  Put a 67k in series with a 100K and bingo, I got -38 at about center travel of the pot.   I measured the pot alone and got 4.88.  This with the 47K still connected to the 1st lug on the pot.  Could it be the 15k resistor which limits the pot and the 47K is the negative feedback?

If you want less -ve bias voltage, you either need to decrease the bottom 'lower leg' of the voltage divider or increase the 'upper leg'. (The 'knee' is the output of the divider - the pot wiper). So that amounts to a decrease in the (47k) tail resistor, or an increase in one of the resistors (e.g.: the 15k) above the knee of that voltage divider.

A voltage divider basically a way of lowering DC voltage by tapping off an output somewhere along the length of resistance between the highest potential (where you are connecting the input voltage) of the voltage divider and the lowest potential (ground potential). If we call the portion of resistance between the voltage input and the tapping off (output) point 'R1', and the portion of resistance between the tapping off point and the ground 'R2', then the formula for a voltage divider is Vout = [R2/(R1 + R2)] x Vin. In this case Vout is measured at the pot wiper.

In this case its easier to experiment with the 'R2' resistance, because you can solder-tack other resistors in parallel with it (the 47k) to lower its resistance without having to remove anything from the board. Tacking another 47k in parallel will halve the resistance here, but you may need to decrease it even further. Its a case of suck it and see.

[Ed_Chambley]

So I should go back to the 220k bias dropping resistor and make adjustments to the 47k, the lower leg of the voltage divider.  No problem in doing this.  What I am not clear about is how you come up with the tube wattage at idle.  I understand the percentage of plate dissipation, but how do you get wattage measurement?

The only thing I have ever done is install a 1ohm resistor on pin 8 to ground and measure mV.  I have done the shunt method, but usually I install bias tip jacks and put the adjustment pot next to them on the back of the chassis. I have seen and read where this is not the best way to go, but it seems to work well once the amp setup correctly.
 
With my plate voltage being 462, so 70% that would be 37.8mV.  What would be the problem with changing the bias dropping resistor to get within the necessary range as opposed to changing the lower leg of the voltage divider?  If I understand this correctly, it is just lowering the voltage prior to the diode.

[sluckey]

What I am not clear about is how you come up with the tube wattage at idle.  I understand the percentage of plate dissipation, but how do you get wattage measurement?

Multiply plate voltage times plate current. Here's a very good write-up on biasing your amp...

http://aga.rru.com/TechTips/disp.cgi?file=LV/BiasAdj.tt

What would be the problem with changing the bias dropping resistor to get within the necessary range as opposed to changing the lower leg of the voltage divider?

Either method accomplishes the same thing. I prefer diddling with the 220K. So does Doug. His Plexi layout even labels the 220K as the "Bias Range Resistor". See pic...