diytube.com

I was wondering if it's possible to use a "plate to plate" feedback scheme like the RH-84 uses on 6L6, EL34, etc Clementine builds? If it's simply done by adding a RFB resistor you could add it between pin 3 of the output tubes to an unused coupling cap hole.

Anode follower feedback works well in power amps. It's covered somewhat in RDH3 and 4 if you want to get down and dirty with the math

Cheers!

I like the idea. I must say that I love the RH84 amp that Tom built for me. How about a Chubby Clementine using 1625 and 12SL7? From what I read you either high mu, or high transconductance for amp works perfectly.
So do you think that the 12SL7 can push low impedance node through 1625 grid without distorting? Over my head for sure. Brian

The "RH" plate to plate feedback works well with most tubes.

A rule of thumb is to use a feedback resistor of approx 4.5X to 5X the driver stage's plate resistor value, for a 100K resistor, use 470K, that's more than close enough. For 47K, use 220K, etc.

THis NFB scheme, like all others, will reduce the overall gain of the amplifier, but this isn't usually a real problem in most cases.

Even though we call it "plate to plate" feedback, it is actually plate to control grid feedback for the power tube, the connection just ensures that the interstage coupling cap is included in the circuit.

HTH

/ed B

In the early design phase of the Clementine (exactly a year ago), I was planning on using this feedback scheme and a pentode tube as the voltage gain tube. But my linearity tests of my test amp sucked, and the pentode (6SJ7?) needed to be biased 'just so' for low distortion. I bailed at that time for the trusty, linear 6SL7. I also bailed on the feedback technique due to the low plate impedance of the triode (and RDH4 recommending against it).

Shannon

brians7 wrote:I like the idea. I must say that I love the RH84 amp that Tom built for me. How about a Chubby Clementine using 1625 and 12SL7? From what I read you either high mu, or high transconductance for amp works perfectly.
So do you think that the 12SL7 can push low impedance node through 1625 grid without distorting? Over my head for sure. Brian

The 1625 is an 807 with 12.6V heaters, basically.
This means that you can apply DIRECTLY the RH807 schematics to the 1625.

BUT, you will not be able to use the 12SL7 (or 6SL7) as driver, since it does not have enough transconductance.

On the other hand, the ECC81/12AT7 family can be heated with 12.6V, so there is nothing stopping you to build an RH807 with the 1625 (except the wish to use a 12SL7 at all cost ).

Not to mention the fact that you can switch the ECC81 for a 5965, or E180CC with very interesting results...

While there is debate on whether the 1624 can be seen as a DH substitute of the 807, I recall having induced my friend Elvis to convert his RH807 (the one shown on the site) to 1624s...

He did it, and the sound was very similar - but a conversion to DC heaters was necessary as well, since hum was quite loud on his Lowthers

Anyway, the RH807 can be built with 1624 as well...

A further mod would be the RH-307A (VT225).

I have actually never built the RH807 since I had neither the tubes nor the sockets and caps. In the meantime I have managed to find some 307A tubes, and would be willing to build the RH307A, but am still missing the sockets and caps...

I believe the final schematics would be similar to the RH807... but have yet to write a spice model for the 307 in order to optimize the values if necessary.

Just a word of warning - do not build the RH807 as shown. I've got an issue with the resistor in the supply to g2, I solved it by using two stabilizer tubes and then using a mosfet for regulation while having a couple of hefty resistors going to earth in case the current is reversed. (the issue with large resistors to g2 is mentioned in the STC datasheet for the 807).