So what is the basic order of operations for assembly?
The beginnings are obvious,
- Raw supply
- B+ regulator
- B- regulator
- Measure/match CCS mosfet
...but it's about there that things get fuzzy. What would be the next logical step?
The beginnings are obvious,
- Raw supply
- B+ regulator
- B- regulator
- Measure/match CCS mosfet
...but it's about there that things get fuzzy. What would be the next logical step?
Stuff signal circuit boards. Then test to catch errors and match FETs with tube sections. Then assemble, wire all the boards together, hold your breath, and pray there's no smoke.
I tried to write down some of my confusions and issues in posts 224. Of course, I don't have your experience, so many of mine were rookie mistakes.
Also, I put the JFETs on sockets for testing and was glad I did. Since mine were Toshiba black, the low current required some fiddling.
Jac
Last edited:
I'm a rookie as well, I haven't built one of these!
(And honestly, I'm a card-carrying member of the "Fake it 'till you make it!" club, I count myself lucky if I can properly identify a resistor from a capacitor 3 out of 5 times.)
I've read your posts a few times, you've earned a gold star for doing it first and good documentation. It's very appreciated. 😀
Last edited:
Sy, any reason why this could not swing the volts needed to drive some compaction output tubes? You mentioned using the topology in some of your power amps.
In the EO article it's mentioned that the stage would be an excellent driver for output tubes! I think it can swing 40 as-is, and probably more as the plate voltage in increased. And with the follower, any input capacitance of the output tubes would be trivial to drive.
Stu, is that right?
Stu, is that right?
That's what I am thinking. Just wanted to confirm because we are moving from mV to V. Didn't know if it would run out of steam at higher levels or whether or not it was simply a function of headroom in B+.
In all designs, SY strives for complete freedom of overload - if you read all his articles, it becomes clear that that particular aspect is a huge pillar of his design language. (And very fast recovery if there is an overload.)
Anyway, the most straightforward way of doing that is to make it able to swing many more volts than you would ever call upon it to do.
I'd love to see a power amp made with this as a driver!! I'd make one except that I have to build my Red Light District first. (I have all the parts) 😀
Anyway, the most straightforward way of doing that is to make it able to swing many more volts than you would ever call upon it to do.
I'd love to see a power amp made with this as a driver!! I'd make one except that I have to build my Red Light District first. (I have all the parts) 😀
Well summarized 6L6. I am considering some 6cb5a or 6hj5 tubes with this FE. In triode, both make for a lovely output tube....and cheap!
This is exactly what I was mulling, esp as I have balanced outputs to feed into it. But decided I should actually get of my fat ar5e and do some sums, which of course I haven't yet...
6hj5 needs about 40V swing to drive to clipping, so with a bit more headroom, I think it should work very well.
Yeah, use a higher B+ and tubes which will swing more volts (e.g., 6CG7) and you'll have an excellent driver stage.
I am covered in rue- I have PCBs for you which I keep not taking to the post office. I promise to remedy that as soon as I get back to town.
I am covered in rue- I have PCBs for you which I keep not taking to the post office. I promise to remedy that as soon as I get back to town.
No worries here. Life is busy and this just happened to pop into my head when thinking about a PP amp I was planning.
Sy, what are the parameters to look for in a good cascade tube? Anything unique or anything to avoid
Cascade or cascode? If the latter, top or bottom?
I'm sure you know this, but the EO is neither- a lot of people think "cascode" when they first see it, but the topology is actually quite different, as is the performance.
I'm sure you know this, but the EO is neither- a lot of people think "cascode" when they first see it, but the topology is actually quite different, as is the performance.
Just RTFM, right
. So the jfets are operating as source followers driving the lowZ input at the cathode of a grounded grid tube? Is the grounded grid acting as I/V stage loaded with a CCS?
. So the jfets are operating as source followers driving the lowZ input at the cathode of a grounded grid tube? Is the grounded grid acting as I/V stage loaded with a CCS?
Nope, it's voltage driven from the followers- they see a high load from the CCS (CCS impedance plus plate resistance divided by mu, which is still large). The cross coupling is what gets you the enhanced PSR and CMR. I thought it was pretty straightforward, but Ian Hegglun assures me that it's actually a new and novel circuit.
Just keep in mind that tubes are stupid and only know what grid to cathode voltage is, not grid to ground or cathode to ground.