Looking for Mikael's KT88 Schematic

[porkchop61]

Thanks Mikael, I thought this might be the reason, less is better sometimes.
I was just listening to my amp (just finished it), and it sounds wonderful! Thank you so much for sharing the design. I built a stereo version for expense reasons.
I've added the switch for UL, Triode, Pentode selection. Boy, big increase in power output in Pentode mode!

Is it just me, or do the power tubes run hot The rectifier is cooler than the power tubes
I'm more familiar with push pull designs, so I assume it's because the single tube runs full power all the time. Does anyone run forced air cooling, or is convection enough for this application? I was thinking that I could add a small power supply type fan pulling air into the base that the amp is mounted to. This way it would force air out of the cooling holes around the power and rectifier tubes.

I'm also glad I have 25W power resistors on the cathodes of the power tubes. They get quite warm, even though I calculated the dissipation at around 5 watts.

Glenn

 

[Mikael Abdellah]

The power tubes should definitely handle the heat with convection cooling as far as I know.

About the resistor.. yes they get VERY hot! When I started looking for a good bias point I used a 30W rheostat.

 

[porkchop61]

Okay, thanks.
I tend to at least double wattage ratings on power resistors.
Cooler lasts longer
Glenn

 

[porkchop61]

Okay, I just finished this amp (stereo version), so I thought I'd post a few pictures (amp porn).
Thanks to Mikael for posting this design in the first place, and to everyone that was patient enough to put up with all my questions.
I'm VERY happy with the sound of a SE amp, I'm used of building PP designs. No crossover distortion with the SE
This design also gives me a great platform for tweaking here and ther to see what effect it has on sound. It's pretty "punchy" with the EL34's in there vs the KT88's. I'm going to order some JJ 6L6GC's to also try.
Now I just have to dig out my Altec A7 VOC's to give the amp a proper voice
Thanks again to all.
Happy New Year!
Glenn

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

 

[Mikael Abdellah]

Wow! really nice pics. Happy new year!

 

[porkchop61]

I'm trying to calculate bias for the EL34 tubes I'm currently (Ha) listening to.
It's been a while since I've done this, so I had a few questions on where to measure voltages from for calculations.
Is the plate voltage measured with respect to ground or to the cathode of the tube? Do I use the B+ measurement in the calculation, or the plate voltage?
I'm using the calculator at:
Weber bias calculator

In UL mode I get:
pin #3-8 375v
pin #5-8 -29.8v
pin #4-8 378v
pin#8-GND 30.4v
B+ is 418v (a bit higher this morning)

In Triode mode I get:
pin #3-8 376v
pin #5-8 -29.4v
pin #4-8 376v
pin#8-GND 30v
B+ is 418v

In Pentode mode I get:
pin #3-8 374v
pin #5-8 -30.2v
pin #4-8 382v
pin#8-GND 30.6v
B+ is 414v (dropped a little on this measurement)

I'm using the standard 470R bias resistor called for on the schematic.

Any help would be appreciated.
Thanks
Glenn

 

[porkchop61]

No one?
I'm just questioning if I should use the actual plate voltage (plate to cathode) or the actual B+ (plate to chassis). According to the calculator at Weber's web site, it's the actual plate voltage.
I'm currently running EL34's in UL and using the calculator I get 23 watts and 63mA, so this is just under the rated 25 watts for the EL34.
I just want to make sure it's the actual plate voltage used in the calculation.
Thanks
Glenn

 

[Mikael Abdellah]

Really wish I could help you but I have no idea. Sorry...

 

[porkchop61]

That's okay, I'm pretty sure it's the actual voltage from the plate to the cathode.
Glenn

 

[Depanatoru]

You have to use the voltage between plate and cathode . It's the plate voltage to ground minus cathode resistor voltage.
In fact it's simply the Ohm's law , tube power dissipation (W) = voltage across the tube ( plate to cathode ) (V) x plate current (A) .

 

[porkchop61]

You have to use the voltage between plate and cathode . It's the plate voltage to ground minus cathode resistor voltage.
In fact it's simply the Ohm's law , tube power dissipation (W) = voltage across the tube ( plate to cathode ) (V) x plate current (A) .

That's what I thought, I was measuring from plate to cathode.
Looks like I'm just within the limits for the EL34. Probably won't last too long at that bias point. It does sound good though!
Thank you

Glenn

 

[jmillerdoc]

Anyone have a copy or link to the schematic of this amo that uses the feedback from plate to plate with the 33k resistor?

 

[scott17]

Anyone have a copy or link to the schematic of this amo that uses the feedback from plate to plate with the 33k resistor?

Post #5 on this thread:
KT88 SE Abdellah (modified by Alex Gendrano) + AK.. - AudioKarma.org Home Audio Stereo Discussion Forums

 

[revintage]

The silliest thing to do would be to add plate-to-plate feedback to this driver. It will heavily increase distorton due to the lowZ load on the driver. If you for some reason still want to try, LEDs in the cathode circuit is also totally useless. Must be an unbypassed resistor to help things up. Unfortunately triodes and plate-to-plate does not go together very well. Best done with a potent pentode driver.

If you want lower Zout instead of the very high 4ohms of the UL version you should use cathode feedback over the outputwinding as George suggests.

 

[arold19]

When I look to my James 6123 transformer, I understand that if I use the 5K primary winding, it not have 43% screen but 35%... Is that important?
The 43% screen is available only when using the 3.5K winding.

Thank you.