Quick quick question re Series resistor power handling

[emenasche]

I want to up the value of the cathode resistor for a pair of EL84s with what I've got on hand, and I'm wondering if this will work:

Schematic calls for 100/2 watt ohm—in some versions, 150/2 watt in others. The old resistor that was there was showing about 180, and it sounded okay. I've got a 100/5w and it's a little too crunchy for my ears, so I was wondering if it's safe to stick, say, 20 to 50 ohm/1 watt in series between the 5-watt and ground? Or am I better off grabbing a couple of 220 1 watters and running them in parallel.

The amp is a Kalamazoo Reverb 12 that's showing about 305 volts B+ for the power tubes...

Thanks

schematics linked below. The first one seems closer to the amp I have, at least in its original form. I made a bunch of changes to the preamp/reverb but nothing from the PI on.

http://www.rru.com/~meo/Guitar/Amps/...2-schem-bf.gif

http://www.rru.com/~meo/Guitar/Amps/...Reverb12-a.gif

[eschertron]

Hi! Welcome to the place. From your post, I'm assuming you have sufficient experience to know how to work on high-voltage devices. Regardless, BE CAREFUL!

The 100R/5W in series with the 20R/1W will split the power dissipation proportionally to their resistance, in other words, they will each dissipate max power at the same current. So they'd be a good match-up. But 120R in a circuit where 180R sounded good may not change the sound enough to please you. The reason I'm saying that is I've been playing around with a 2xEL84 amp of my own, and have gone from a cathode resistor of 120R up to 300R and back (I think 150R is in there now) to get 'that sound'.

If 2W resistors lasted *any* amount of time in that amp, you could string all three 100R+20R+50R together and the 50R would be OK. You could experiment with different combinations, and when the magic cathode resistor value is found, order and install a 5W wirewound resistor of the desired value for longevity. I'd advise keeping an eye (or IR thermometer) on the 50R resistor if you do use it, while testing your changes.

[emenasche]

Escherton, thanks very much for the welcoming words and for the info: this is just what I needed! I really appreciate you taking the time.

[The Dude]

Firstly, welcome to the place!

I would first have to ask if we know what the bias current is. "Sounds a little too crunchy" could be for a number of reasons that have nothing to do with the cathode resistor. IMO, you should change cathode resistors to achieve proper bias rather than some tonal issue.

[emenasche]

Thanks, and of course you're right—though the change in sound came after I replaced the old 150-ohm resistor, which had drifted high to about 187.

Even if I calculate the bias, I'd still want to chain some 1-watt resistors off the 100/5w, at least for now until I can source resistor of the right value that can handle the power. So if that's safe to do, I should be good to go?

Firstly, welcome to the place!

I would first have to ask if we know what the bias current is. "Sounds a little too crunchy" could be for a number of reasons that have nothing to do with the cathode resistor. IMO, you should change cathode resistors to achieve proper bias rather than some tonal issue.

[eschertron]

How are you measuring the tube dissipation to get a 'proper' bias?

With the amp at idle, you can measure voltage across the existing 100R cathode resistor, and calculate the current through the resistor. And while the current will drop if you add more series resistance, you could use that value (double or triple it for an engineering margin) to estimate max dissipation across your 1W candidates. Still good to go?

Fire could be a real possibility if the resistor is not an acceptable type and overheats. You'd need to be very attentive to the amp during experimentation like this.

[Gingertube]

For Your Info:

The shared 100 Ohm resistor results in the EL84s being run VERY hot. This is the same bias level as the original AC30 which use a common 50 Ohms for 4 off EL84.
That is too hot for some modern production EL84 although JJ seem to cope.


100 Ohms shared is the same as using separate 200 Ohms, one on each EL84.
When using a shared cathode bias resistor, matching of the output tubes becomes critical, otherwise one tube can hog most of the idle current and it then overdissipates and expires.
I strongly recommend using separate bias resistors and bypass caps for each EL84. Matching of tubes then is not crictical and amp reliability is enhanced.

The recommended individual bias resistor for EL84 HiFi work is 270 Ohms.

The shared 150 Ohms option discussed (same as 300 Ohms individual) seems too cold for guitar amp.

The last blown up AC30 I restored, I rebuilt with separate bias resistors and bypass capacitors on each EL84.
I used 2 x 470 Ohm 2 Watt metal film resistors in parallel (for 235 Ohms) bypassed by 47uF/50V on each EL84.
That ran the EL84s at a more conservative (safer) power dissipation - this is equivalent to a shared between 2 tubes resistor of 118 Ohms.


The owner of that amp (an original 1963 Copper Front panel AC30) was very happy and said that he could not tell the difference sonically from the original.

Cheers,
Ian

[SoulFetish]

If the voltages on the schematic are correct, biasing those EL84s at 50mA will cause you to go trough output tubes so quickly. Plus, that 2W resistor will fail. I'm not sure what the P-P load impedance is, but if it's around 8K (common for EL84s), you probably want to bias the grids at -11/-12V. Plus, put 1k current limiting resistors on the screen grids to protect them. This design runs the screens almost at plate voltage