I used 2500 ohms for the maximum power tests where I cranked the power supply to max, 650 volts. I used the 1250 ohm Plitrons with the 8 ohm load on the 4 ohm tap. These transformers are rated for 400 watts at 20 Hz, so they made full power at any frequency fron 10 Hz to 40 KHz. At 250 WPC the tubes were dissipating far more than spec, but not glowing. You wouldn't want to leave them going at 250 watts continuously. Music has at least a 10 db peak to average ratio (usually at least 20) so you could turn it up till it clips and nothing would melt except for your speakers and ears.
I also tried them with 3300 ohms, and still got about 150 watts which my "80VA" OPT's didn't like. Plenty of power at 1KHz but severe distortion and red plate tubes at 50 Hz.
Smaller sweep tubes with a lesser peak cathode current rating can't handle the 2500 ohm load very well.
George, why would you get severe distortion and red plate tubes at 50 Hz with a 3300 ohm xfmr?
Last edited:
Core saturation at 50Hz will result in loss of inductance in the transformer and screw the impedance.
That about explains it all. The impedance isn't what makes the difference, the size of the OPT does. The 80VA transformers were designed for guitar amp use. The lowest note on a standard electric guitar is about 80 Hz. If I run these transformers down to 20 Hz I can put about 20 to 25 watts through them before they saturate. I can crank 150+ watts through them at 1KHz. This is normal for all OPT's. When a transformer saturates its inductance drops to a very low value. This makes them look like a short circuit returning all of the audio output power to the tubes to be turned into heat. Redness appears quickly. At a higher frequency the OPT can handle much more power.
This is why the ratings for many OPT's are meaningless. The best yardstick for true power handling at low frequencies for OPT's is weight. Note the scales are different for P-P and SE transformers due to the gap. A 5 pound P-P opt like the above mentioned Schumakers is good for about 25 watts. A 25 watt SE opt weighs about 10 pounds. So when you see those Eastern Audio SE OPT's on Ebay that weigh 6 pounds each and claim a 50 watt rating, realize that they will pass 50 watts at 1 KHz, but they are a 10 watt OPT.
That is one open question for me as well...what primary impedance to use with 6HJ5s running at 450V B+.
!!!!!I have an important question!!!!!
I was discussing plate caps for the red board in another thread when the conversation went like this:
This prompted an unsolicited email from Pete offering me a free PC board if I will actually build an amp. I have several unfinished projects, but how can I refuse a deal like this. So, what do I build? I have tested amps using this board that made anywhere from 15 WPC to 250 WPC. Like anything in life there are a few rules:
I'm sure that Pete would like to sell a few more boards, so a design that no one could copy but made a zillion watts is really not helping anyone. So, we need to choose something that YOU the forum reader would want to build and actually could. This means that oddball parts, or modifications to the PC board aren't acceptable here. Adapdations to fit tubes (octal) that are more available world wide may be considered if it doesn't turn out to be too difficult.
My current financial situation doesn't support expensive parts purchases like transformers right now. It is expensive supporting two households on one salary. This build must use parts that I already have.
So the power transformer will be one of several Anteks that I have, one or two of the "10 pound" transformers depending on power level, or one of several Hammond / Allieds that I have. Combinations of two or more Anteks is possible as long as the design can be duplicated by a reasonably competent builder.
The choice of OPT gets slimmer. I have plenty of the 6.6K guitar amp OPT's seen in many of my red board experiments, but they impose some power limitations. They will work well at 3300 ohms. I have a pair of "Handwound Transformers" of questionable quality. They are 6600 ohms and will support about 100 WPC. They will not work at 3300 ohms, too much HF rolloff. They are similar in size to the 100 Watt Edcors so I could swap the OPT's at a later date if desired. I have the monster Plitrons, but I don't want to commit them to a "low powered" design.
Tubes. 6HJ5's work well for up to 100 WPC. 35LR6's will work to 200 WPC I have plenty of both. 6GV5's are probably good for 50 to 60 WPC. I have 20 of them but haven't tried them in the board yet. I have limited supplies of several other sweep tubes that plug in to the board.
Octal tubes don't plug directly into the board but I will fabricate an adapter if there is sufficient interest. 12 Pin compactrons aren't common outside of the USA. Octal sweep tubes will work in the board and there are several choices with similar pin configurations. My board currently has E130L's in it right now which look promising, but I haven't listened to them.
I am leaning toward an amplifier using the 6600 ohm OPT's. The power output will be dependent on the available B+ voltage. This is where a mild deviation from Petes plan may take place. The Antek power transformers have two HV secondaries. One can be bridge rectified and fed to the board to supply the screen and driver plate supplies using the on board regulators in the prescribed manner. This requires adding a second set of diodes to make the bridge. The second secondary will be bridge rectified off board and added in series with the first supply to make 600 volts or so to really crank some sweep tubes. A small transformer will be needed to create the negative rail. This same technique can be used with two of the "10 pound" transformers.
I believe that an amplifier can be made in this manner that is reproducible, reliable, and appeals to a world wide audience. What do you think? Would you build one? If so which one?
I'm sure that there are a few that want to see that zillion watt monster, even if they wouldn't build it themselves. OK, If this one works out, and once I get my other projects out of the way, I'll make one. This isn't it.
Greetings George....
What if the choices were octal or compactron and then power output in perhaps 25 watt steps. (ex 25w, 50w, 75w, or 100w). I would be more than willing to help fund part of this effort. Please let me know what I can do to help.
Ray
What if the choices were octal or compactron and then power output in perhaps 25 watt steps. (ex 25w, 50w, 75w, or 100w). I would be more than willing to help fund part of this effort. Please let me know what I can do to help.
Ray
George...
This sounds like a great project. Initially, I bought 2 boards from Pete, thinking I might build another one with some "changes" ...not that I'm unhappy with the performance; the DCPP is a wonderful amp.
This sounds like a great project. Initially, I bought 2 boards from Pete, thinking I might build another one with some "changes" ...not that I'm unhappy with the performance; the DCPP is a wonderful amp.
I've got a fever, and the only prescription is more power (cowbell)!
I was thinking of an simple mod to reasonably squeeze more power from ye red board. A simple recipe: 450v B+ and bigger OPT's. If I'm not mistaken all that is really needed is a more resistive R21?
I've thought to myself if only I could squeeze powerdrive in there... That would take some offboard work.
600v B+ would be awesome.
I was thinking of an simple mod to reasonably squeeze more power from ye red board. A simple recipe: 450v B+ and bigger OPT's. If I'm not mistaken all that is really needed is a more resistive R21?
I've thought to myself if only I could squeeze powerdrive in there... That would take some offboard work.
600v B+ would be awesome.
Last edited:
If there is enough interest, I'm not adverse to spinning a version of the board that has octal sockets... and making some other changes, within reason.
Food for thought...
Pete
Afternoon Pete,
When you get a chance, please check your email. I sent you "some food for thought"!
Ray
When you get a chance, please check your email. I sent you "some food for thought"!
Ray
The first "within reason" change would be to bring the heaters of the input tubes and the output tubes out seperately. They could share a common pin so only one extra connection is required. Most available input pentodes are 6.3V. Sweep tubes came in all different voltages, and the oddball flavors are cheap, some are on the dollar menu.
I thought that too. In fact I tried it once but only got some fried mosfets. I may have connected things wrong in a hurry. I think that PowerDrive could squeeze a bit more power out of this board, and possibly improve the overload / transient recovery. On the other side of the coin, if you build it to a 100+ WPC level, you should never see overload! PowerDrive? Maybe, but it would add to the complexity. I'll try it eventually. Now that I don't have to worry about destroying my only board, maybe I'll try it again. With luck, I may even have a three day weekend with 3 rainy days!
450 volts, why limit yourself. My board has seen 650 volts, and produces 250 WPC. How much power do you need? I think that two of the 10 pounders will be near 700 volts. R21 can be increased but you will get to the point where it dissipates too much power. Hence the split power supply concept. Which brings me back to another simple mod. Add two more diodes in the main power supply, and maybe the negative supply so that simple toroids without CT's can be used in FWB. The extra diodes are DNP with a conventional transformer.
An octal version of the board would be more useful world wide, but which octal tube set do you support? There was a list a few posts back of a few tubes that could be used with one pin configuration. The venerable 6BQ6 and 6DQ6 are good choices too since the 6BQ6 has been sold for under $1, and their pinout is compatible with the 6L6GC. Of course I like the 6LW6 because of its sheer power capability, but it has a very odd pinout. I don't think that all choices can be handled on a common layout though.
What about an "octal" version with jumpers around the socket to jump for different tubes?
Ray
Ray
Well, firstly I think you should go for more power. <50WPC tube amps are a dime-a-dozen and you have proven that this amp responds well when the knobs are turned up. I have done it myself and was quite impressed. If someone wants less power, they can just stick with Pete's design. If they want a bit more power they can just increase B+ without really changing anything else (including the tubes).
At first I was thinking that it should just be a 6HJ5 amp, but that is already sort of an old hat. I'm making one to run at 450-500V B+ and so are one or two other people. Mine hopefully won't require any special off-board components. Initial testing with an Antek AN-4TK400 was promising. If Pete is willing to rev the board, then I don't see any reason not to switch to an octal socket if it makes the board more flexible.
At first I was thinking that it should just be a 6HJ5 amp, but that is already sort of an old hat. I'm making one to run at 450-500V B+ and so are one or two other people. Mine hopefully won't require any special off-board components. Initial testing with an Antek AN-4TK400 was promising. If Pete is willing to rev the board, then I don't see any reason not to switch to an octal socket if it makes the board more flexible.
...and if you are going to rev the board, Pete, it might be good to open up more possibilities for the B+ filter caps. Some of the locations are too tight to fit anything else...especially the 10uF caps. I was able to stuff 500V caps in for the big ones, though a few more mm around some of them would have made it a little easier.
Good point. I could remove all the B+ caps and R21 do a doubler as you suggest. 600v and screen supply would be easy with a big Tek power transformer. Hmmm.
I still have a full assembled board sitting here, just haven't gotten around to firing it up. I have it modified to run the 6HJ5s as well.