Speaking of safety, I thought I'd mention to anyone think about building a hot-rodded red board some alternative parts for the Shade network:
220k, 3W, 1kV working, 2kV peak
680pF, 1.5kV, mica
or
820pF, 2kV, ceramic
220k, 3W, 1kV working, 2kV peak
680pF, 1.5kV, mica
or
820pF, 2kV, ceramic
The first time I flipped the switch on my board I ran 6JN6's with a 6600 ohm OPT and 350 volts of B+. I got 30 WPC, so you should be in that range with an unmodified board and a 6.6K OPT.
I went back through the entire 700+ posts to find that I have not tried that exact combination. I saw 50 WPC with 450 volts and a 6600 ohm load on the 6JN6(post #139), 67 WPC at 500 volts, 72 WPC at 550 volts and 80 WPC at 600 volts (post 149). The 6JN6 could not drive a 3300 ohm load even at 450 volts(post 153). The 6GV5 is similar in power dissipation and cathode current capability to the 6JN6.
I also saw 125 WPC with 520 volts and a 3300 ohm load on the 6HF5. I could assume about 100 WPC with 450 volts and a 3300 ohm load using a tube with the capability to drive a 3300 ohm load.
It looks like to be in the 75 WPC range you need a 6600 ohm load and 600 volts of B+ using just about any sweep tube, or you need a 3300 ohm load and a 450 volt supply, but the tube needs peak cathode current capablilty in the 1 amp range and 24 watts or more of plate dissipation capability. it may be possible to use the 6GV5 and 450 volts with a 5000 ohm load, but I have not tested this combination.
The red board that I have is not in my lab right now. It is being used as a real amplifier shaking the house walls. A second board is being constructed using the information discovered in all of the experiments so far. When it is done the two boards will trade places and the experiments will resume. I have some 6GV5's and a way to make a 5000 ohm OPT, but no means to test the combination right now.
You know the resistors that are in my board and are still alive.....They are these BC Vishay resistors. I got them from Digikey about 10 years ago, they no longer carry them. I got to get some more. I mistakenly thought mine were off value. I found out that the cheap Harbor Freight meters give wrong numbers when the battery gets low. The meter works, but reads wrong when the 9V battery has 6 volts in it. I did find a fried screen resistor. It has been replaced and the two channels are now identical. Board #2 is being populated, but I'm missing a few parts.
Russ,
Did you ever characterize the impedance of those OTs you are using? The 8417 tube in the GE manual lists 4.2K primary for 100 Watt output, which you quoted earlier for their history. It's getting confusing comparing the various quoted B+ voltages versus power lately without the primary Z quoted also.
Don
Did you ever characterize the impedance of those OTs you are using? The 8417 tube in the GE manual lists 4.2K primary for 100 Watt output, which you quoted earlier for their history. It's getting confusing comparing the various quoted B+ voltages versus power lately without the primary Z quoted also.
Don
If I use 6GU5 inputs, 6LF6s with the AS-4T400 transformers, one per channel (~560V, right?), separate 320V input power supply, and Citation II output transformers (3.2k later version), is it reasonable to expect to get ~150WPC at mid frequencies? I know the bass will be somewhat limited by the output transformers.
Going to attempt monoblocks, so can't use the board.
Thanks,
Stuart
Going to attempt monoblocks, so can't use the board.
Thanks,
Stuart
Yes! This has happened to me. They read the voltage wrong too. Drove me batty until I figured out it was the meter. Oh well, what do I expect for three bucks.
I haven't done much with them other than clean them up, repaint them, and put them back on the amp that they came from (which I am trying to restore). I had lost the voltage ratio measurements that I took, so I retook them last week. They have 3.3k primaries and an 8 ohm secondary with a 4 ohm tap. I think I was confused about what they were before and my dad said that those number rung a bell. He still thinks they are 100W transformers, but I don't think they have enough inductance to deliver that much power at 20Hz. I suspect 60W is more realistic, but I didn't spend much time with the red board on these. He doesn't know what they came out of (they were surplus).
I'm currently leaning towards more of a compromise for the DCPP 6HJ5 setup. Edcor has a 4.2k primary 100W PP transformer that will give me a little less power but better distortion figures.
Thank you for your very thorough answer, much appreciated. I'll keep collecting parts for now and leave the transformers for the last. By the way, Vacuumtube.net let me know they only had 3 6GV5's for me, so I'll have to change plans a bit.
I have already sketched out some ideas for my chassis. Will post here when there is progress to report.
"By the way, Vacuumtube.net let me know they only had 3 6GV5's for me, so I'll have to change plans a bit."
VacuumTube.net lists 6GY5 for $2, and that has the same pinout as the 6GV5. Its rated at 18 Watt diss. (versus 17.5 Watt for 6GV5) but has a considerably bigger plate (has the same size plate as 6HJ5) than the 6GV5. They also list 21GY5 for $1 and 17GV5 for $2.
VacuumTube.net lists 6GY5 for $2, and that has the same pinout as the 6GV5. Its rated at 18 Watt diss. (versus 17.5 Watt for 6GV5) but has a considerably bigger plate (has the same size plate as 6HJ5) than the 6GV5. They also list 21GY5 for $1 and 17GV5 for $2.
Last edited:
Both ESRC and Vacuumtubes.net have 6HJ5's in stock for $4. They have no plate cap and are proven to work well in this amp up to 125 WPC. I have found several good tubes listed for cheap at Vacuumtubes.net but when I called them they are not in stock. I called them this morning for another tube they had listed for $2 but there was no stock. I confirmed the stock of 6HJ5's at both places this morning, but did not check on the 6GY5.
A minor modification is needed to run 6HJ5's in this board. It involves the placement of 8 resistors and no trace cutting is needed. Rknize posted a picture of the modification somewhere back in this thread. I'll post another when I get that far with my builds. I ordered the parts needed for me to build 2 more red boards from Mouser yesterday. Assembly is already started.
Last edited:
So while tube lab is finalizing everything so i can start my parts order for a 6hj5 with the 2 10 pounders for power supply (I kinda wanted a top cap one with nice big red leads, i have lots of wire from the tv shop i worked at), is it safe to order the edcor 6-8k transformers for a 100 watt build?
I am sure that a tube with a plate cap can be found. It seems that everyone else wants no plate cap.
The big Edcors come in several load impedances:
3300 ohm, 4200 ohm, 5000 ohm, and 6600 ohm.
I get 125 WPC with a 3300 ohm load using two 10 pounders for power.
I get 75 WPC with a 6600 ohm load using two 10 pounders for power.
These are known data points. So for sure you will get 100+ watts with a 3300 ohm load. I assume you will get 100 watts or more with a 4200 ohm load. I will guess that 5000 ohms will be just short of 100 watts, and I know that 6600 ohms will get you 75 watts.
I can rig up a 5000 ohm load once my second board is built. I have a round about way to get to 4200 ohms. These tests are probably a week or two away if there are no interruptions. So If you want 100+ watts using 6HJ5's from two 10 pounders, based on tested data, you will need 3300 ohm transformers. If you aren't in a big hurry I will have more data in the future.
If you want a tube with plate caps I know that any of the 30 or 35 watt sweep tubes like 6JS6C, 6LR6, 6KD6 will get you 125 WPC with two 10 pounders. They are not cheap. The 6HF5 will get you to 100+ watts with a 3300 ohm load (tested). I think the 6GV5's may get you there with a 4200 or 5000 ohm load, but I haven't tested it yet. ESRC has some 6GV5's in stock but they are $4 each.
I am going to a hamfest tomorrow and Saturday. I stop at ESRC to find more sweep tubes for my experiments. More data will be forthcoming as it develops.
I had been using 6DK6s as my drivers for the past few days and today I switched them out for 6FG5s. Can't say I noticed anything different but at least there are other options out there on the dollar menu.
I have been having issues getting the drivers to balance, there seems to be about a 20 volt difference between the two tubes on each side. I may try swapping the two lower value tubes to one channel and the higher reading tubes to the other. I think I was getting readings of 140 and 119 for the left and 115 and 137 for the right.
Also just to note I measured the board B+ 288volts, 144v on the screens, -61v bias and 600 on the plates.
I think I may look through my boxes of tubes and see if there are any 6GU5s to try. I have had my speakers hooked up full tilt a few times and it is pretty damn loud with no real audible distortion.
I have been having issues getting the drivers to balance, there seems to be about a 20 volt difference between the two tubes on each side. I may try swapping the two lower value tubes to one channel and the higher reading tubes to the other. I think I was getting readings of 140 and 119 for the left and 115 and 137 for the right.
Also just to note I measured the board B+ 288volts, 144v on the screens, -61v bias and 600 on the plates.
I think I may look through my boxes of tubes and see if there are any 6GU5s to try. I have had my speakers hooked up full tilt a few times and it is pretty damn loud with no real audible distortion.
Last edited:
The 6GU5 gives more gain (volume) without any more distortion as long as you don't push the amp into cliping.
I plan to stop in at ESRC if I can arrange it over the weekend to look for more test subjects. Unfortunately he will be at the hamfest as long as it is open.
Some have suggested the 6GY5. I got some waiting for me. Any other ideas?
I, too, noted the large size plate on the 6GY5s I acquired in a speculative buy. I plan to use them for some nefarious purpose. One thing holding me back is the PITA skirted mini plate caps used on all those squat Compactrons w/caps. The Chinese plate caps available as new stock are generally a poor fit for these and the caps of capable soldiers like the 6BQ6. I have a solution involving a standard fuse clip with silicone tubing to dress it up, but I haven't tried it in a real design as of yet. The concept is a fuse clip of appropriate size with a 1/4" fast-on termination, a covering of silicone tubing, and a squirt of hard silicone to seal the top of the tubing and properly center the wire. I generally use the limp silicone insulated wire sold for model electric cars. It's available in some attractive colors (though the yellow could be better), it's very flexible, and also holds up to the heat.
"Any other ideas?"
21JZ6 or 6JZ6, 21JV6, 6KE6, 21HB5A, 21HB5 or 6HB5, 42KN6 or 6KN6, 21LG6A, 6LB6
the -KN6 come in two versions apparently, dual tube sections in parallel or a later single section design. ( I wonder what existing tube they used two of to make it originally, sounds like it was some quick and dirty hack job until the real design could be finished)
I think the following are all equivalent functionally to the 6HB5:
6KE6 21KA6 21JV6 6GY5 21HB5 (and other filament V's)
and 21HB5A and 21JZ6 are equivalent to each other functionally, but slightly different from the 6HB5 or 21HB5 non A (not counting filament V's or caps). I don't think there ever was a 6HB5A, but the 6HB5 and 21HB5A have different curve sets on the GE datasheets.
-----------
small plate caps........ Ugh, yeah I got some 1/4 inch fuse clips too.
21JZ6 or 6JZ6, 21JV6, 6KE6, 21HB5A, 21HB5 or 6HB5, 42KN6 or 6KN6, 21LG6A, 6LB6
the -KN6 come in two versions apparently, dual tube sections in parallel or a later single section design. ( I wonder what existing tube they used two of to make it originally, sounds like it was some quick and dirty hack job until the real design could be finished)
I think the following are all equivalent functionally to the 6HB5:
6KE6 21KA6 21JV6 6GY5 21HB5 (and other filament V's)
and 21HB5A and 21JZ6 are equivalent to each other functionally, but slightly different from the 6HB5 or 21HB5 non A (not counting filament V's or caps). I don't think there ever was a 6HB5A, but the 6HB5 and 21HB5A have different curve sets on the GE datasheets.
-----------
small plate caps........ Ugh, yeah I got some 1/4 inch fuse clips too.
Last edited:
I think it needs a bit more gain. My younger brother currently lives with me was listening to it and he noted that my Simple SE sounded louder when listening to songs we normally listen to.
OH, I forgot to mention that the 6/21HB5 series have the least compatible pinout with Pete's red board. Better to get one of the equivalent tube versions to try out on it. So 6GY5 and 6JZ6 say, would cover the bunch.
Leaving 6/42KN6, 6/21LG6A and 6LB6 as the only different ones.
Leaving 6/42KN6, 6/21LG6A and 6LB6 as the only different ones.
Last edited:
Yes, Pete never imagined that the big dumb blonde one would take his board and extract up to 250 WPC from it. THe 6CB6 provides enough gain to get from a CD player to the level needed to drive a little sweep tube to 20 WPC.
There are several ways to boost the gain. The simplest is to stuff in an input tube that has a higher transconductance. The 6GU5 seems to boost the gain a bit without causing other problems. Increasing the plate load (I now have 39K) will also increase the gain. So will increasing the Schade feedback resistors. All of these tricks will lower the plate voltage on the input tube, so going too far will move the tube into a non linear region.
I observe that my board the way it is set up now doesn't have enough gain to hit clipping on most CD's. It is still extremely loud, far louder than a SSE. I will figure out how to extract some more gain when I have two of these to play with.
Long term I am looking at a new design with 3 total stages from CD player to speakers, but that is still a non functioning breadboard.
I'm off on a 500 mile road trip. Be back with more tubes tomorrow night.
"Long term I am looking at a new design with 3 total stages from CD player to speakers, but that is still a non functioning breadboard. "
The 9DX base video amp tubes would be a good fit for the front end. Triode front end, pentode drivers, with sweep outputs. The 9DX base offers 25 different (and mostly cheap, yet high performance) tube types on the same socket, for the ultimate tube rolling. Almost never see them in DIY amps, yet they are always on the sales lists. They make the usual suspect audio tubes look like CK722's. I'm using 9DX for my build. Easily extended to experiments on g2 drive as well.
The 9DX base video amp tubes would be a good fit for the front end. Triode front end, pentode drivers, with sweep outputs. The 9DX base offers 25 different (and mostly cheap, yet high performance) tube types on the same socket, for the ultimate tube rolling. Almost never see them in DIY amps, yet they are always on the sales lists. They make the usual suspect audio tubes look like CK722's. I'm using 9DX for my build. Easily extended to experiments on g2 drive as well.