Just Built Pete Millett's "Engineers Amp"

It's a nice little amp--nice build, cgnman59! I installed a 500K input level pot on mine, but it was too high in resistance, so replaced with a 50K audio taper pot and made sure it was insulated from chassis ground. Doing very well now and very happy with the sound. Considering the 50W monoblock version as well.
Dave
 
Nice!
Something I wouldn't think of
(TV tubes for audio amps)
I was always under the assumption if they were good for audio, someone would have used them already
 
TV tubes are great for audio as long as you're listening to the sound and not the stamp on the side of the envelope.

The disadvantage, of course, is the higher output impedance requires a different winding ratio for the output transformer.

If you think about it, the TV is a grueling environment. High voltage, high current, medium gain, lots of hours of life. The tubes are inexpensive and readily available. What's not to like?

Just don't tell anyone about the high performance characteristics and low prices, or the prices will go up.
 
Dave451, thanks I am impressed with your chassis. I am hoping to build another cabinet from walnut and replace the current one that I made from left over oak flooring. The top plate for mine came from Front Panel Express as specified by Pete Millett. This is my first real experience with a tube amp. My previous amp was a Pioneer SX-D5000 which I purchased in 1982 with Pioneer HPM-700 speakers. I thought it was a nice system. But to my ears this 18 watt amp blows it away. The question on my mind has been; how does the DCPP amp compare with commercial tube amps?
 
The other "problem" with TV tubes is the oddball sockets required and sometimes the strange voltages involved for heaters. Of course new ones aren't being made but you can usually assemble enough stock of old ones to get by.
 
Cgnman,
Comparing the DCPP to my vintage commercial amps, it holds up very well, I think. I ran Pete's analysis for my own unit and it matched his measurements very closely, meaning it is accurate and particularly strong in bass response for its power. There is very little frequency response roll off at either end with the feedback engaged and the sound is 'crisper' to my ears compared to, say, a Scott LK-72. It might sound fairly 'bright' into certain speakers. I miss having tone controls and, as I said, I did add a volume pot up front, but the signal path is pretty simple for this amp, which helps with clarity. Bass response is very tight. I have listened to it for extended periods without fatigue. As far as comparing to newly available tube amps, I couldn't say, having not tried any.

Gadget, yes the compactrons are a little strange, but you can get the 12 pin sockets on ebay (as I did) and the tubes are still available for relatively little money. Someone earlier said that they may have gotten 'bought up,' but I found them readily available. Speaking of odd voltages, I came into a cache of old TV and audio tubes and there were indeed a lot of compactrons with 30, 18, and 13 V filaments, among others, as well as 6 and 12V versions. There are some neat ones like the 6AC10, which is a triple triode that I think Pete uses in some other designs.
 
The question on my mind has been; how does the DCPP amp compare with commercial tube amps?

My old Jolida 302b continues to collect dust, and I thought I had it pretty well sorted with NOS tubes and other mods. With stock Edcor iron, the DCPP amp beats it no problem.

jeff
 
I built a version of the DCPP six years ago, and use it every day. It is the best sounding amp I ever built, and I have not built another amp since getting it done (started one but stalled). My version is point to point wired, with 12 tubes (nothing succeeds like excess): That is two damper diodes to rectify, four tung sol 3cb6s (odd heater voltage tubes are cheaper), four 6GV5 output tubes with caps, and two 6sl7s that serve as the four feedback resistors. As John Broskie said, the Schade style feedback makes Pentodes sound like Triodes. I used some old Heathkit output transformers and a honkin EICO power transformer, and a bunch paper in oil capacitors in the sound circuit and motor runs in the rectifier part. The thing weighs a ton, and looks steam punk, but sounds great. My all in cost was $210 then (if memory serves - half of that for the Heathkit output transformers), but it would have been smarter to save hours of work and buy Pete's circuit board.
 
I built a version of the DCPP six years ago, and use it every day. It is the best sounding amp I ever built, and I have not built another amp since getting it done (started one but stalled). My version is point to point wired, with 12 tubes (nothing succeeds like excess): That is two damper diodes to rectify, four tung sol 3cb6s (odd heater voltage tubes are cheaper), four 6GV5 output tubes with caps, and two 6sl7s that serve as the four feedback resistors. As John Broskie said, the Schade style feedback makes Pentodes sound like Triodes. I used some old Heathkit output transformers and a honkin EICO power transformer, and a bunch paper in oil capacitors in the sound circuit and motor runs in the rectifier part. The thing weighs a ton, and looks steam punk, but sounds great. My all in cost was $210 then (if memory serves - half of that for the Heathkit output transformers), but it would have been smarter to save hours of work and buy Pete's circuit board.

:needpics:
 
I need another project like I need to pay more taxes... but I think I'm going to start collecting parts to build one of these. So thanks a lot.

I built a Millet Starving Student headphone amp (12AU7 version) a few years back. Still using it to this day. His designs are spot on.
 
I mention John Broskie. He refers to the "Schade" style feedback (the internal plate-to-grid feedback used in the DCPP) as "partial feedback." For a pretty deep dive into this design feature (about which the somewhat cynical Broskie is very enthusiastic), read his lengthy 2001 article (in it he also explains using a tube as a virtual resistor, which I have done with 6SL7s), found here:

https://pearl-hifi.com/06_Lit_Archive/14_Books_Tech_Papers/Broskie_John/2001/01_02_Mar_2001.pdf

One obvious question is whether the tubes-as-feedback-resistors reduces distortion, increases it, or increases it preferentially as 2d harmonic. I don't know - I don't have the equipment to test, other than ears.

My amp falls far short of beautiful - I assembled it as a proof of concept then attempted to gussy it up a bit, but rough edges are easy to see. Nevertheless, pics attached.

Millett Amp w 6SL7 Virtual Resistor for Feedback - 12 tubes.jpg Millett amp w 6SL7 virtual resistors for feedback view 2.jpg
 
Amazing build, Perseo! I considered doing point-to-point but opted ultimately for the 'big red circuit board.' Never heard of using a tube as a feedback resistor, so I'll certainly have a look at the article you linked.
 
I mention John Broskie. He refers to the "Schade" style feedback (the internal plate-to-grid feedback used in the DCPP) as "partial feedback." For a pretty deep dive into this design feature (about which the somewhat cynical Broskie is very enthusiastic), read his lengthy 2001 article (in it he also explains using a tube as a virtual resistor, which I have done with 6SL7s), found here:

https://pearl-hifi.com/06_Lit_Archive/14_Books_Tech_Papers/Broskie_John/2001/01_02_Mar_2001.pdf

One obvious question is whether the tubes-as-feedback-resistors reduces distortion, increases it, or increases it preferentially as 2d harmonic. I don't know - I don't have the equipment to test, other than ears.

My amp falls far short of beautiful - I assembled it as a proof of concept then attempted to gussy it up a bit, but rough edges are easy to see. Nevertheless, pics attached.

View attachment 1215764 View attachment 1215767

I like the birch corners!
 
Thanks. You guys are nice. It shows me as a new member but I posted on various forums in the past, perhaps not this one. I have been inactive in the hobby for a while.

Birch is my go to.

The idea of a triode functioning really as a resistor is found here and there. A Japanese designer dubbed a circuit using this as a load as a "super triode." Broskie's Aikido essentially uses a triode on top of a triode as a driver and the same arrangement with the bottom triode as a load in a cathode follower output. My attraction to it was that it should be a very good quality replacement for a mere resistor (using a high quality resistor in the cathode to bias the thing and get the triode to have the right dynamic resistance value) and, in sharp contrast to old time designers of commercial tube amp equipment, I am happy to use as many tubes as possible rather than as few as possible.
 
Back
Top Bottom