trinityamps.com

I want to use EL34s in my amp and I put the 3.5K ohm striped wires from the OT to pin 3 of the output tubes and switched out the 620K resistor to the suggested 1M resistor and was only able to get 10mv for max bias. I had a second 1M and added it in series and for it to 20mv. If that trend continues I’ll need a 4m resistor to get the 35-40mv for bias. That is pretty far from what the builders guide says but it also says “you may have to play with the bias resistor to get it into range”.

My question is to those of you using EL34s what value are you using and what have you biases the 34s to?

Thanks
Mark

Seems to me that you are going the wrong way.

Adding a 1M resistor in parallel with the 180K reduces the resistance to 152K. Adding 2M in parallel makes that 165K and 4M would make it 172K.

The stock 680K that's switched in parallel for KT66s makes it 142K. But if you need more bias than the KT66 setting can provide, the resistance should be less. The comparable resistor in a Hiwatt DR504 running two EL34s is 110K. The circuitry is somewhat different, but to get in that range the resistor in parallel with the 180K should be around 270K.

I don't have a Triwatt, so this is pure speculation. But if you try using smaller resistors than 680K, start with the bias pot on its lowest setting and watch for red plating. Like the builder's guide says, you are doing his mod at your own risk!

Hey thanks Mitch for the quick and detailed info. I have put it back together the way it was with the Sovtek 6V6GTs. I don’t have the resistors to continue on so for now it will be the 6V6GTs.

Still asking if anyone here has used EL34s in their build and what have they found that works

Thanks
Mark

It would be worth trying EL34s for an even more authentic sound.

If you have some 1M resistors, two of them in parallel would be 500K, three would be 333K and four would be 250K.

By the way, for those who don't know, the formula for parallel resistors is 1/[(1/R1) + (1/R2) + (1/R3) .... + (1/Rn)].
Just take the reciprocal of each resistor, add them up, then take the reciprocal of the sum.

So I went ahead an ordered some 1W metal oxide resistors. 470k,330k,270k,and 220k. I can start with the 470k and see how it acts and go from there. I do have 4-1m metal oxide but would prefer not having a stack of them in the amp. I would think changing out the 620k for a “single” lower value would be and look better. Thanks for your help Mitch.
Mark

Yes, using a single resistor to replace the 680K is the best way to go and you don't want to wreck the neat look of your board and wiring. I'm not a fan of scabbing together resistors in parallel or series, either, if a single standard value is available. What I had in mind is to temporarily tack solder in a few of your 1 megs just to see if that gets you the bias range you need. You could do that while you wait for the resistors you ordered to arrive.

Hey thanks Mitch, the amp is back together I’ll get back to it soon not in a hurry. Do want to see what the 34s will sound like. I’ll post my results.

Still asking about others that have used EL34s in their Triwatt
Mark

The resistors came, so I connected the 3.5k taps to pin 3 and removed the 620k resistor and alligator clipped in the first one which was 470k and the max bias was around 5. So then I tried the 339k and was only able to get like 1.6 max. I stopped using lower values and tried the 1M and it’s max bias was 12, I added a second 1M in SERIES for 2m resistance and it went to 21 mV max, I tried a 3rd 1M in series and it only went to 23 mv. That should have gone to at least 30mv and it only went up by 2. I stopped there. Don’t want to hurt the amp so I quit. The bias isn’t reacting like it should be. Oh pun 3 voltage stayed around 465vdc. It didn’t move much. I did this last night and slept on the results so they may not be exact but very close

I want try the 34s but need the bias to be in line with the tubes. What should be my next move
Thanks Mark

I did some voltage checks that are on the layout and all of them were very close and some were even exact. Here are 3 that were off

V1 pin 6. 100vdc. Layout spec 120vdc
V4 pin 6. 294vdc. 236vdc
V4 pin 1. 308vdc. 274vdc

Just a note we have used 6V6GT and KT66 tubes in the amp and they bias great.
Mark

It would be interesting to hear from somebody who has actually tried using EL34s in a Triwatt.

Just speculating again, but it looks like reducing the value for the 180K feeding the bias circuit doesn't work, so maybe you have to increase it. Adding any resistor in parallel with the 180K can only decrease the value.

You could try replacing the 180K with a 220K and removing the 680K (or whatever you currently have in its place) and see if that gets the bias moving in the right direction. Start with the bias pot at minimum setting and keep an eye on the B+.

Ok Mitch, check it out. I had bias switch in the 6V6 position 180k still installed and got
V5 24.0mV
V6 21.5mV
That’s what I was seeing with 3 1m resistors in series, looks better to me. The builders guide says the bias switch in the KT66 position is 15-18v higher on pin 3. Do the EL34s like it at 465vdc better? Or is the 450vdc good?
At 450vdc on pin 3
Gotta go to work, I’ll switch out the 180 for a 220(is metal oxide ok)?
Thanks Mark

So here is what I got. Put in a 220k in place of the 180k with the switch in the 6V6 position and the max bias was
V5 41 ma
V6 36ma
Which I think is good but the voltage on pin 3 was 436vdc. What do you think about all that?
I’m thinking Stephen put higher voltages on the KT66 and EL34s for a reason can I get that back closer to the 465 where it was before all this?
Thanks Mark

I’ve been thinking, if there is not a resistor in the 620k position, can’t I just flip the bias switch to the KT66 and bring the voltage up. I know I may need to increase the resistor value again, say to 270k for the range.
What do you think?
Mark
Edit: if that is the case would a 2w resistor be in order?

You should run the EL34s at the higher voltage. That's why I suggested removing the 620K resistor. That way you can use the KT66 switch position to get the higher B+ without reducing the 180K resistor's value. You may still need to change the 180K to a higher value to get the bias in the 35mA range. 1-watt resistors should be OK, but if they get warm you can go to 2-watt.

Hi Mitch thanks for the quick response. I tried the switch in both positions and as the voltage went up the ma went down. Now that we have determined by changing the 180k resistor to a higher value and switch on the 6V6 position I was able to get it to the 35-40ma range. So If I up the voltage to the KT66 setting I have to raise the resistor value to get the bias range higher with the higher voltage. (As voltage goes up current goes down) I think all that sounds about right!
Mark
Sorry Mitch I totally missed the part about removing the 620k resistor. It’s out now but I haven’t moved the bias switch to the KT66 position but I will tomorrow.

Ok I finally got it right. Pin 3 is 455vdc at 40ma. Bias pot range is 30ma-45ma. Biased to 38ma. Didn’t crank it because it’s 1100pm. Bias resistor is 220k metal oxide. Thanks Mitch for your help we are exactly where I want to be.

Thanks again for your help Mitch!
Mark

Thanks for the update, Mark. Glad I could help a bit. I should have realized earlier that the total resistance would have to go up rather than down to get the bias you need, but it's the negative voltage that had me a bit mixed up.

Yes, the plate voltage goes down if you increase the bias, so it can take a bit of fiddling to find the right setting. Some people set the bias so the power EL34s draw idling is 17.5 watts. That's 70% of the 25-watt maximum dissipation for an EL34. You are at 18.2 watts now (455 volts x .04 amps). Other people just tune it by ear, which is fine as long as you don't get the bias too high. Hotter bias tends to sound better, but it shortens the tube's life.

Hope you get a chance to give the EL34s a good workout and let us know how they compare to KT66 and 6V6. My guess is they will be louder with a richer sound.

Sorry for being a bit thick but I'm confused by this.

I thought the EL34 adaptation was to select 6L6 on the switch, short out the Zenner, select the impedance above your needs (16 Ohms for an 8 Ohm speaker) , re bias if needed and Bob's yer auntie ?!
Speaking of which, can you run EL34's with a 16 Ohm speaker ?

The Triwatt now has 2 sets of primary taps for the power tubes 3,5k taps for EL34s and 8K taps for KT66s. I wouldn’t run 6L6s in this amp. As far as speaker ohms it’s not the tubes that determines that it’s the OT secondary taps. This amp has 4,8 and 16 ohm speaker selection.

The 6V6 setting provides more bias range. But the B+ is lower because of the Zener diode. That's why you would have to short the Zener out. But the bias range might not be enough without increasing the 180K resistor as we found.

The speaker and the impedance ratio of the output transformer determines what impedance the output tubes see. Connecting an 8-ohm speaker to the 16-ohm tap would reflect 4K back to the 8K primary winding. That's close to using an 8-ohm speaker, the 8-ohm tap, and the 3.5K primary winding.