Hello doug......yes, I chased the dragon of the diamond coated filamentless tubes since oct. 1992 when an issue of Scientific American had an article on diamond film semiconductors, and as an aside, described a prototype diamond tube built by some MIT techs....I found a company in england that will coat anything you want in diamond coating, and investigated getting some thin copper rods coated for testing . Further checking into the claims of the original S.A. article, I became convinced the effect wouldn't be strong enough to provide current for anything but a small voltage amp tube, such as a 12AX7, and never something as heavy of a current as an output tube...so I dropped the matter a couple years ago. The ICFRD is basically a connection from the cathode of the driver tubes, in this case 12BH7s, to the cathode of the opposite output tube cathode rail....with a 1K resistor, as I recall...I stumbled upon this by hooking up the driver cathode to the same side cathode rail, and a nice stable feedback ensued, turning the output/driver stage into an oscillator. Reversing the connections to the opposite sides (hence the inverted part) was the trick. I had read that negative feedback in a gain stage lowers output impedance, and with this circlotron setup with the 6550s, I was happy to register a drop in the output impedance with my rather crude test setup for determining such. Although there is a formula of mu and plate current (I forget the exact details) which gives an impedance result, I used a variable power resistor, with a stable white noise input to the amp, then adjusted the resistor for maximum heat....so the test was pretty hokey, but I think should have some validity. I sensed max heat at 225-250 ohms. What is the nano tip device you make? Medical equipment? I am curious...edd
Last edited:
It would be interesting if you could make measurements of the output into a dummy load, to see what output power you are achieving. I would worry that with such a high impedance load as you are using, you are going to be pretty limited with 150V HT supplies in the circlotron output stage.
The absolute maximum peak output voltage you could possibly get is going to be 150 V minus whatever is dropped across the bank of tubes that is conducting during that half of the cycle. But to be anywhere near a linear response, the peak voltage would probably have to be a lot less than 150. To get 40 W into 192 ohms you'd need 124V peak, which would seem highly unrealistic. If you went to 256 ohms you'd be even worse off, needing 143V peak for 40W.
Chris
yes, I am likely off on my guestimate of power but the thing is very loud and remains clean as I crank it up. A guy in columbia had a distortion meter but it was not working and after several attempts to get him to fix it up so I could measure things accurately, I gave up. Having worked on and owned medium sized amps of around 40wpc, it seems to have a similar head room in actual use. I am happy with the performance, and into other things right now. The 192 ohms is figured with the rated speaker impedance, the actual measured dc resistance is 175 ohms per stack. Presumably that lower figure would have some bearing on the matter...if I went to 256 ohms (rated imp),that would be 233 ohms dc resistance per stack. I think the impedance match would be closer, wouldn't that maximize transfer of available energy and raise overall efficiency? That is what I will try and see what happens to overall sound levels, as I port in and then short across one more set of speakers as soon as I can find one on craigs list and do the re-wiring to series connections...edd
hello ,
I wanted to make out a will with tubes EL / PL 504 in mode, little as l ampli Graaf GM 100, tubes had been to rise in mode Triode but the power below 1 Khz falls seriously, the conso on l had 100 V= with only 2 tubes EL 504, food was 160 mA.
J wanted to make out a will with regard to the joined plan, what is that this assembly little to be more successful?
I have 10 tubes but I need env. 160 V = so that ca is correct?
I wanted to make out a will with tubes EL / PL 504 in mode, little as l ampli Graaf GM 100, tubes had been to rise in mode Triode but the power below 1 Khz falls seriously, the conso on l had 100 V= with only 2 tubes EL 504, food was 160 mA.
J wanted to make out a will with regard to the joined plan, what is that this assembly little to be more successful?
I have 10 tubes but I need env. 160 V = so that ca is correct?
Attachments
That design looks really inefficient!
You might want to read this thread:
http://www.diyaudio.com/forums/tubes-valves/161112-what-tubes-tube-amp.html
You might want to read this thread:
http://www.diyaudio.com/forums/tubes-valves/161112-what-tubes-tube-amp.html
what is a "source line cabinet"? Do you mean take all the drivers out and built custom cabs for them? I had considered that earlier...the two quads of each cab are pointing at rather large angles away from centerline....and getting all the cones closer together and at a closer angle would allow a more linear sound field....but those sturdy little cabs allow for a decent enough sound for now...I do think i have overestimated the rms power on the unit. Until I can really check the thing correctly, I will guestimate I have 20-25 wpc especially since the tubes are aging and I should get them on the tester to see how they are holding up after a year of use in a circuit they were not intended for.....I presume they are still ok as I don't bias them as hot as the ones in a typical push pull amp, such as the mac 275 I was using....right now the bias is -8vdc grid to cathode which brings the unloaded B+ from 175 down to the 150 figure...edd
the true vertical arrangement may give a sharper imaging, ala the "2 slit experiment" but I feel the close circular arrangement would max the low range functioning more as a single surface. I wouldn't mind experimenting with another complement of speakers and do an A-B comparison....then running them in parrallel and on top of the standard box cabs, that would be interesting with about a 10 ft overall height. I actually took the whole thing out to kansas and played it in a field...it held up pretty good, and that was before I reduced the neg bias from -16 to -8vdc...the pwr went up considerably after I got it back home and reset it... I was being deliberately conservative in the beginning with not wanting to ruin all those brand new JJs. The main problem with things now, is that I moved into a house with a very echoey living room, and havent had time to correct that yet. It sounds perfect for a small amped guitar though...
"What is the nano tip device you make?" The owner who started the company had a boring job with RCA at the end of the phonograph days, that is when CDs were first introduced by Memorex. He realized that the "cantilever" design indeed had a future. Dragging a stylus across atoms, molecular structures while reflecting a visible laser off the back to "profile the a substrate". ...so it was HiFi that still makes me a paycheck! Talk about <dragging it on & on>
The "fred")): Interfrometric common mode A-B cancellation perhaps. Wow. I have been extremly busy and still have not had the chance to study your "FRED",, but am excited if indeed you are doing A-B signal processing.
Last night my washing machine felt that it was a good day to die (*Chief Dan George), I wish it had some tubes inside, (did they ever?) Still reading.
Doug
The "fred")): Interfrometric common mode A-B cancellation perhaps. Wow. I have been extremly busy and still have not had the chance to study your "FRED",, but am excited if indeed you are doing A-B signal processing.
Last night my washing machine felt that it was a good day to die (*Chief Dan George), I wish it had some tubes inside, (did they ever?) Still reading.
Doug
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.