• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Valve Mosfet Hybrids: a solution

Status
Not open for further replies.
Hi everybody.
Some time ago I had written a thread asking for advice when interfacing the low ampage ability of valves with Mosfets. Those require quite a bit of current to load/unload their Gate capacitance.
I had tried all sorts of ways which little success. I had tried most of the designs available on the net and the result has been invariably very poor. It seems this point is ignored ...

After lots of tries and errors, I have a good solution and would like to share it here.

First, my feeling is that the simpler is the better, also for sound.
Therefore, I use only one triode for V amplification, and a pair of Mosfets for transforming the V into the amps needed for the spkrs.

I like using an integrated pair of complementary mosfets (ALF06NV8) because they are matched already and made for audio purposes.

The solution I use is: to use a Mu follower configuration for the triode, the trick is to use another Mosfet, low power high voltage (and low G capacitance) instead of a valve in the Anode circuit. I started from the paper I found on the net "the hybrid mu follower" (google will bring you to it).

Now, for the interfacing, the problem is double:
You need enough stability to have a stable adjustment of the Idle I and of course no voltage at the output (offset null).
This is done with two adjustable Rs, one between both Gates and one to the + Vcc.
You adjust both, starting with the R between Gates set to 0.
When their nominal value is too high, you do get the proper values but they are not stable: they fluctuate at +/- 100 mV, maybe up to volts. All wrong.
I got around the problem by using low values for those adj: between Gates only a 1K and to Vcc some 5 K.
Now go back to your design and just try to use low values: you will be able to get a +/- 3 mV offset on the output.

The result is the best hybrid amp I made. As it is, you can easily swap the triode for another type, to choose the best for you. Quite straightforward.

I hope this will help, it took me lots of time to arrive to a solution. My 50 watts amp has only two active parts and no feedback. Square signal up to 130 kHz, using ECC88. Wonderful sound, now pinpointing I have to work on my spkrs cabinets (back horns).
 

Attachments

The large, complete schematic runs off the page within the Word doc itself, and even in landscape orientation you will not get it to fit unless you reduce it.

If you save the Word Document as a normal Web page (in the save-as options), the images will be placed into their own folder.
 
Brovo V1

Your idea has been implemented now 🙂 the Bravo V1
 

Attachments

  • 310-354_HR_0.jpg
    310-354_HR_0.jpg
    344.5 KB · Views: 439
Here is the schematic !
Sorry for the delay but I had to tidy it, adding some variable R's to make it fully adjustable if you want to try another tube.

It works great and is probably the simplest you can have.

Even if the schematic looks crowded, it remains simple. In fact there is everything, including the Mu follower polarization stuff.
 

Attachments

Last edited:
DN2540 cascode works very nicely too and is very simple to do as well.. you can even use a different FET on the bottom...

ecc812 is a French mazda? I have never seen triode curves, but if you use a ccs then I suppose you can easily adjust the current flow.

How does it sound with the cathode un-bypassed? Ever consider LED bias?
 
The Tower schematic shows there is a V offset at the output … it is supposed to be compensated by the TL071. Better avoid the offset rather than compensate for it. Just imagine the TL dies for some reason …
On the other hand, bipolars have their distinctive sound patterns (third harmonics), I prefer to avoid them altogether.
Class A anything will sound great, in that case you can go and see all the Zen from N Pass, they are also supposed to be good. I tried some and found them in my hearing to be not worth the effort. But it is a question of tastes, of course.

If you don't have an ECC88, try the excellent 5963, very neutral and giving a balanced sound. its gain is slightly inferior to ECC88. The ECC chisels out perhaps the trebble, not reinforcing them, but they seem to be a bit ahead. I will try adjusting on my tweeters, which have a higher sensitivity than the wide bands they complement. Should be fine.

If you won't I can try to post the PCB and layout.
By the way, the 'PSW' are electronic fuses, called Polyswitch.

Ah, last thing: this designs absolutely hates a short circuit. There is no short circuit protection.
The shortest one will probably kill you Mosfets.

Have a good time.
 
Status
Not open for further replies.