KT66 power tubes in Hotrod Deville

[Micahstephens838]

Hey people question about biasing KT66 tubes in my HRD 4x10.
I got a new set of Genelax Gold Lion Kt66 tubes to replace the 6L6s(was advised this would be okay by the place I bought them) the problem is with my bias adjustment pot all the way down, I'm getting a reading of 104mv at my bias test point. with a plate voltage of 470v-480v on the hotrod and the tube rating at 25w, I'm positive that these are running too hot and i cannot adjust it any lower. BUT I played through it regardless for about 30 seconds and I loved the tone of the tubes. Is there a way to Mod the bias circuit in order to accommodate the new tubes, or maybe I'm missing something. Thanks all!

[JazzGuitarGimp]

Increasing the value of R83 will allow the bias pot to go more negative. I'd start with 100K and adjust as needed from there. The schematic says the bias should be adjusted to -54.6V for 6L6's. It would be nice if you could find a value for R83 that will allow you to get up to, maybe -50V at the top end, down to a little more negative than what you need for the KT's at the bottom end. That way, you can go back to 6L6's without having to break out the soldering iron again. If you are not able to find one value for R83 that will get you the range you need to achieve this, you can increae the range of the bias pot by lowering the value of R77, though I wouldn't change R77 if you don't need to.

[Micahstephens838]

Increasing the value of R83 will allow the bias pot to go more negative.

Hey thank you so much for your response! I have this posted on the technical discussion forum also and some people recommended me lowering the value of R76 to get a lower voltage out of the bias supply. Can you explain the difference of what you are saying and what they are saying? I really appreciate it!

[JazzGuitarGimp]

You're welcome! I just duscovered you have a more actice thread on this in the Technical Discussion subforum. That's what I get for starting at the For Sale subforum and working my way up!

[Micahstephens838]

You're welcome! I just duscovered you have a more actice thread on this in Garage Talk. That's what I get for starting at the For Sale subforum and working my way up!

I submitted that response prematurely on accident haha. I had some questions

[JazzGuitarGimp]

My first thought was to reduce the value of R76, as others have suggested. This will cause the bias circuit to draw more current which, I believe is unnecessary. Increasing R83 will get you the same net effect from the standpoint of getting the bias supply output more negative, and it will slightly reduce the current through the bias supply.

[Micahstephens838]

My first thought was to reduce the value of R76, as others have suggested. This will cause the bias circuit to draw more current which, I believe is unnecessary. Increasing R83 will get you the same net effect from the standpoint of getting the bias supply output more negative, and it will slightly reduce the current through the bias supply.

Got it. So what is the original value of R83? I'm seeing 51k(just want to make sure we are seeing the same thing).

[JazzGuitarGimp]

I am seeing the following:

R76 10K
R83 51K
R82 B25K
R77 100K

[Micahstephens838]

The schematic says the bias should be adjusted to -54.6V for 6L6's. It would be nice if you could find a value for R83 that will allow you to get up to, maybe -50V at the top end, down to a little more negative than what you need for the KT's at the bottom end.

Okay how do those bias reading translate to mV. I'm using the bias test point on the Hotrod deville

[JazzGuitarGimp]

It's a bit of a moving target. Read Note 5 on the schematic: the specified voltage (-54.6V) is for reference only - it will get you in the balloark, but it's going to be different for each set of output tubes you put in it. The more important number is the current through the tubes. Note 5 says to set the bias for 60mV at the bias test point. This equates to 60mA, since the test point voltage is being measured across a 1-ohm resistor. Keep in mind that this measurement tells you the current through both output tubes, so divide by 2 to get the current through each tube. Then multiply the voltage at the plate of the tube by the current flowing through it: 475V x 0.030A (30mA) = 14.25W divide this nuber by the tube's max plate dissipation (30W for 6L6GC) for percentage of bias: 14.25 / 30 = 0.475 (48%). Keep in mind that the KT66 has a 25W max plate dissipation. Your bias will likely wind somewhere between 55% and 70%. If the bias is too low, you'll wind up with crossover distortion, which is not pleasant. If it's too high, you'll have shorter output tube life. 70% seems to be where most people aim. But I'd say try lower bias settings to see where you like the tone, and if that winds up being less than 70%, you will enjoy longer tube life. If your bias winds up on the cooler side, you'll want to check the tone at full volume, just to make sure you're not getting crossover distortion when pushed (could prove embarrassing at a gig).

[Micahstephens838]

It's a bit of a moving target. Read Note 5 on the schematic: the specified voltage (-54.6V) is for reference only - it will get you in the balloark, but it's going to be different for each set of output tubes you put in it. The more important number is the current through the tubes. Note 5 says to set the bias for 60mV at the bias test point. This equates to 60mA, since the test point voltage is being measured across a 1-ohm resistor. Keep in mind that this measurement tells you the current through both output tubes, so divide by 2 to get the current through each tube. Then multiply the voltage at the plate of the tube by the current flowing through it: 475V x 0.030A (30mA) = 14.25W divide this nuber by the tube's max plate dissipation (30W for 6L6GC) for percentage of bias: 14.25 / 30 = 0.475 (48%). Keep in mind that the KT66 has a 25W max plate dissipation. Your bias will likely wind somewhere between 55% and 70%. If the bias is too low, you'll wind up with crossover distortion, which is not pleasant. If it's too high, you'll have shorter output tube life. 70% seems to be where most people aim. But I'd say try lower bias settings to see where you like the tone, and if that winds up being less than 70%, you will enjoy longer tube life. If your bias winds up on the cooler side, you'll want to check the tone at full volume, just to make sure you're not getting crossover distortion when pushed (could prove embarrassing at a gig).

Okay awesome. Again thank you so much for your help. I ended up changing R83 from 51k to 82k and that put my bias 75mv on the lowest end of the pot. With the equation you gave me 475x.0375= 17.8......17.8/25=.71 (71%). do you recommend me switching the 82k to a 100k to get under 70% or do you think I'm close enough?

[JazzGuitarGimp]

Well, I tend to be biased (LOL) when it comes to bias. I might urge you to take it lower, and work your way up. Again, if you find a tone you like at a lower bias setting, you're ahead of the game in terms if tube longevity. How do you like the tone at 71%? Also, when I said 475V, I was ballparking. The truth is, as you increase the bias current, your plate B+ Voltage will drop, so you need to measure your actual B+ voltage and multiply by that voltage. I'm sorry I didn't say that.

[katopan]

Subtract the couple of mA of screen current and you're just a hair under 70% power dissipation - perfect!

[Micahstephens838]

Well, I tend to be biased (LOL) when it comes to bias. I might urge you to take it lower, and work your way up. Again, if you find a tone you like at a lower bias setting, you're ahead of the game in terms if tube longevity. How do you like the tone at 71%? Also, when I said 475V, I was ballparking. The truth is, as you increase the bias current, your plate B+ Voltage will drop, so you need to measure your actual B+ voltage and multiply by that voltage. I'm sorry I didn't say that.

Haha biased..I get it! Measured my plate voltage at 482 so that bumps me up to 73% I'll switch the 82k for 100k in the morning. Thanks again!

[JazzGuitarGimp]

Subtract the couple of mA of screen current and you're just a hair under 70% power dissipation - perfect!

Katopan is right: since you're measuing current at the cathode, about 10% of that current is flowing from the screen grid, and so it doesn't count towards plate current. So for instance, if you measure 50mA at the test point, whack off 10%, then divide by two, which yields 22.5mA. This doesn't have a big impact on plate dissipation, and since it amounts to a reduction in plate dissipation, I tend to ignore it.