Aleph J for Universal Mounting Spec

6L6

Moderator
Joined 2010
Paid Member
I honestly don't know what is going on with the circuit as a whole, I do know (from adjusting it on my amp) that a change in R27 will change the bias of the output devices, which you can measure by voltage drop across the output source resistors.

Hopefully somebody smarter than I can explain the why that works...
 
Altering R27 is changing Bias, but so little.
No i have R27 at 33K instead of 68K and the voltage drop changed from 550mV to 525mV
This is irritating, so i will end at 10K or what?

so 68K, 47K, 33K next will be 22K, but does that change the current gain and the sound?
I hope i don`t destroy it.
 
Last edited:
As the total of R27 decreases bias will go lower, and this will not affect the current gain. For larger changes in bias swap out the Source resistors. Going 1 ohm would cut the current in half compared to 0R47.

The static current set requires that the voltage across R16//17 + R15 equal the Vbe of Q4, ~0.6V. Lower values of R27 increase the current through R15 and reduce the component of R16//17.

Increasing R18 and R19 will allow you to increase the value of R7, which will increase the open loop gain. As is, you are operating with less feedback than the original, which may or may not be a good thing.

Offset adjustment seems ineffective because you are in the range of 250 ohms to get the output stage the Vge it needs.
 
Oliver, You're looking at roughly 2 amps total bias. That's about where we would expect the amp to be. If too hot, see if you can increase airflow to the amp or increase the size of the sinks.

One thing you are doing that affects the sound somewhat is setting R7 to ~250 ohms. Open loop gain of a differential pair is determined by the load impedance. (That's why you see high impedance current mirrors in so many designs) With vertical mosfets R7 would be in the range of 800R. The net result is less feedback. Distortion numbers will be slightly higher, but whether it impacts the sound is your call.

If you increase R18 and R19 to 1R you will be able to increase to ~440R and restore most of the open loop gain. It might be a worthwhile experiment to see if you can hear the difference once you get your bias settled.
 
Yes i was considering some really huge heatsinks, or some pc fan for airflow.
But the sinks i have got are way too big.

First i wanted to adjust the bias and this appears to be a bigger problem than expected.

Now i have R27 at 10K and the bias is at 406 mV, but i can´t adjust the bias with R8 anymore, it just doesn´t change.
When the R8 resistance gets to low the amp gets unstable and the offset starts moving up and down, or i have like 14 V offset.

With R8 at 2K i can adjust offset with R7 and everything is stable.
So it appears that i could leave R8 fixed at 2K and adjust the bias with R27.

I build Aleph 3, Aleph 30 and Aleph X before, but this never happened.
Of cause i never used lateral Fets before, only in F5 without any problem.
 
R8 ONLY adjusts the bias on the front end in this configuration. If R7 was fixed you could use R8 to adjust the offset, as was done in many previous Aleph versions. Technically you can adjust bias with R8 when the Aleph Current Source is set to allow more current than the output device will pass with the Vgs set by R8 and R7.

Your laterals are a major change since they require a Vgs ~1.2V where verticals need ~4V. That's easier to deal with in the F5 since you aren't balancing a current source and the gain device.
 
Yes i was considering some really huge heatsinks, or some pc fan for airflow.
But the sinks i have got are way too big.

First i wanted to adjust the bias and this appears to be a bigger problem than expected.

Now i have R27 at 10K and the bias is at 406 mV, but i can´t adjust the bias with R8 anymore, it just doesn´t change.
When the R8 resistance gets to low the amp gets unstable and the offset starts moving up and down, or i have like 14 V offset.

With R8 at 2K i can adjust offset with R7 and everything is stable.
So it appears that i could leave R8 fixed at 2K and adjust the bias with R27.

I build Aleph 3, Aleph 30 and Aleph X before, but this never happened.
Of cause i never used lateral Fets before, only in F5 without any problem.

You're going in the right direction by lowering R27...very much lower than 68k as this sets the lower Vgs of the 2SK1530 compared to IRFP240. So if you find that 10k for R27 gets you your desired Output Stage bias current, measure the voltage from Gate to bottom of Source R of the Current Source mosfet this is the Vgs+VR17 say ~2.8V. Now match this voltage with the lower Output mosfet by adjusting R7. You do this by knowing the input stage current thru one of the jfets. If these jfets were matched at 8mA then it's 2.8/0.008 or R7=350 ohms. Set R8 such that it is double this current or 16mA for the input jfets.

Hope that helps...
 
Member
Joined 2009
Paid Member
After drilling and tapping the heatsinks yesterday I wired up my Aleph J this morning and checked the voltages.The p/s is running at +- 22.5V and the offset was at 75mv on both channels which I was able to zero after it had warmed up.Voltage across the source resistors reads 0.55V on one channel and 0.56V on the other;is this a problem as the h/sinks are at approx 47degrees?Should I install the trimmer so that I can lower this?
The amp plays music and sounds pretty good but this is only through a pair of Tandy 4" drivers.
A big thanks to Jim for his excellent build guide
 

6L6

Moderator
Joined 2010
Paid Member
If the temp is cool, you are good.

BUT, do check the temperature of the transistors themselves - pin 2 is the same as the face of the transistor, and verify that they are 65C or below.

Verify that the heat is being removed from the transistors. If it is, everything is fine.