Yeah your right about the filaments. Using two of the AN-2T230, the 230V secondaries could be paralleled, one transformer for the top section, and one for the bottom section. Four 6.3V 4A secondaries would be available to run one 6HJ5 & one 6GU5 off of each. Create an artificial center-tap for the 6.3V secondaries and connect as per the original schematic. Of course still need the bias transformer.
Using a single 200 VA power transformer to run an amp capable of 250 watts of audio output is asking for smoke. The idle disipation of the amp is over 150 watts. It goes to nearly 500 watts at full power. One 2T230 could run the low power variation. Two of them should do OK with the high powered flavor, but I don't have 2 of them so I haven't tried it.
To run this board on two Anteks, parallel the HV secondaries on each transformer and connect one transformer to each bridge. Parallel all 4 heater windings and connect them to the heater terminals on the PC board. Run a jumper wire from one of the heater terminals to the connection for the CT. That connection biases the heaters a few volts above ground. it does not need to be connected to the CT, either side will do. A small transformer is needed for the bias supply.
To run this board on two Anteks, parallel the HV secondaries on each transformer and connect one transformer to each bridge. Parallel all 4 heater windings and connect them to the heater terminals on the PC board. Run a jumper wire from one of the heater terminals to the connection for the CT. That connection biases the heaters a few volts above ground. it does not need to be connected to the CT, either side will do. A small transformer is needed for the bias supply.
So, using one AS-4T400 per channel, for the output plates only should be adequate for a pair of 26LW6?😀
One 4T400 per channel should be more than enough. When I start tweaking on a fresh pair of previously unexplored power tubes I use the target of 65 to 70% plate efficiency for sweep tubes in conventional grid drive in AB1 mode, pentode operation. You get a little better efficiency if AB2 is allowed.
That says if the transformer is only feeding the output tube plates and it is rated for 400VA, you should be able to get 65 to 70% of that power out of your amp. That is 260 to 280 watts. Can you get 260 watts from a pair of 26LW6's? Maybe. I have seen lots of different sized 26LW6's and all of them were at least as big as the 35LR6's that I extracted 250 watts from. Will you push them that hard? For normal use, one 4T400 should power both channels (250 watts total output).
The above generalizations hold fairly well for most P-P amps. And are sized for continuous operation with a sine wave forever. Most people don't listen to sine waves, and music has at least a 10 db peak to average ratio usually more than 20. This means that the transformer must be able to continuously be able to support the idle power consumption, with margin, but doesn't need to be sized for continuous full power operation.
My observations are that an AB1 with occasional AB2 P-P amp using typical audio tubes like 6L6GC's and EL34's will have a best case plate efficiency of 65%. The KT88's may see 70%. Most sweep tubes can do over 70%, some see 75%. Screen driven sweep tubes can do 80%.
So add up all the fixed power consumers like heaters, and driver circuits. Make your best guess about screen dissipation. It is usually 2 to 5 watts per tube. Add all this up in watts. This is the fixed power consumption of everything except for the output tube plates. Then take your plate voltage and idle current for each tube multiply them to get the idle plate consumption. For a simpler computation, just multiply the maximum plate dissipation rating of the output tubes by 4 (4 tubes). Add it to the fixed consumption to get the total amplifier idle power.
Compute the maximum plate power consumption (consumption for both channels deiven to max with sine waves) from the target power output and the estimated plate efficiency. Add that to the fixed consumption to get the maximum power consumption.
Add about 30 watts to these numbers if a tube rectifier is used. Choose a power transformer with a VA rating somewhere between these two extremes. You want to be closer to the maximum power consumption if the amp will be played loud often, used for guitar or other MI amplification, or the owner likes distorted rap music!
That's the $1000 question. Answer is possibly yes on peaks. It's getting the intensity, the dynamics right that would cause me to use that much power, not average listening level. I only care about that in the range above a couple hundred Hz, since I biamp. But my present amp, which delivers 80WPC at clipping is clearly pushed on peaks. My Maggies are only 82dB/W efficient......
Citation II output transformers will be the limiting factor I'm sure.
Stuart
Slamming the amp into clipping on bass peaks will not heat the power transformer much above the idle temperature since the average power requirements are low. Most of the transient energy is supplied by the filter caps in the power supply. Of course having low efficiency speakers will raise the average power level a bit, and if you are going to spend a bunch of time and money to make a big amplifier then the difference in cost between using one big Antek and two big Anteks isn't an issue overall.
The question now is the 4T400 going to give you enough voltage to make the most of the 26LW6's. I will find this out sooner or later since I have 2 of them. I am guessing that maximum firepower will need a 4T450 or a 4T475. I will connect one of my 4T400's up next time I have a chance to work on anything resembling fun. Right now ugly stuff is sucking up my time. Might get time to go to a hamfest tomorrow morning for an hour or two though.
My woodshop teacher is a die hard UK fan. He wants me to come with him to the game tomorrow so I can harrass him all through the game as WVU whupps them again, but I don't have time to spend 8 hours in a car.
The question now is the 4T400 going to give you enough voltage to make the most of the 26LW6's. I will find this out sooner or later since I have 2 of them. I am guessing that maximum firepower will need a 4T450 or a 4T475. I will connect one of my 4T400's up next time I have a chance to work on anything resembling fun. Right now ugly stuff is sucking up my time. Might get time to go to a hamfest tomorrow morning for an hour or two though.
My woodshop teacher is a die hard UK fan. He wants me to come with him to the game tomorrow so I can harrass him all through the game as WVU whupps them again, but I don't have time to spend 8 hours in a car.
I anxiously await your results, George. Even though I have all the parts, I won't be able to start construction for a couple of months.
Hey if anyone is putting in a mouser order, im 6 resistors short thanks to an import error (but hey i got 4 extra other resistors now). Much appreciated if i could papal you the postage and stuff.
6 10k R2W_5083 R28, R35, R38, R40 Resistor, 10k 1/2 watt minimum KOA SPR2C103J Mouser 660-SPR2C103J
Still not sure why the bom says 6, yet only has 4 spots listed.
6 10k R2W_5083 R28, R35, R38, R40 Resistor, 10k 1/2 watt minimum KOA SPR2C103J Mouser 660-SPR2C103J
Still not sure why the bom says 6, yet only has 4 spots listed.
Im tube shopping. Looks like there is a 6CB6A (or its twins the EF190 and 6CF6) that came out in 1940, then the 6CB6 that came out in 1949. What is the diff?
Looks like just 6CB6A's are listed on fleabay.
Looks like just 6CB6A's are listed on fleabay.
I don't know about your dates. The letters after the number indicate a revision that should be backward compatible with the original. The 6CB6 should have come first then it was revised as 6CB6A. I have both data sheets, but one is RCA the other is GE so the comparison may be somewhat invalid. The 6CB6 shows a transconductance of 6200 while the 6CB6A shows 8000. I would go for the 6CB6A.
From what i came up with while looking at several sheets, is the A has a black band in the glass in the center so you can see the plates, and the A has a slow warmup feature for the heater so it doesnt blow as quickly. Both seem to have 8000 transconductance, and some data sheets have both listed as the same thing, only with the A having a longer heat up time.
I made my statement in my previous post based on the information contained on the portable hard drive I had with me at work. I can't look up tube stuff from work or use my work computer to post, I have to use my cell phone.
Now that I am home I can refer to my real paper tube manuals. The 6200 transconductance number is stated in a 1955 RCA data sheet for the 6CB6, but is is tested at 9.5 mA while the 8000 number is tested at 13 mA. That explains the different numbers.
In the early days all TV's and most radios had power transformers. The tube heaters were all wired in parallel and fed 6.3 volts. As with everything, cost reduction eventually chased the power transformer out of TV sets. The tube heaters got wired in series, running directly from the line voltage. Tube heaters have a positive temperature coeficient. Their resistance is low and increases as it warms up. When they are all in series, they better all warm up in the same manner, or the fastest tube will get all of the voltage and light up like a flashbulb. It was decreed that the warm up time will be controlled and specified to be 11 seconds. The conditions for measuring the 11 seconds was specified and each tube got its revision letter upreved when the change was made. This is usually clearly specified in the data sheet.
The 1961 GE datasheet that is on my hard drive does specify an 11 second warm up time, it is not spelled out like it is in other books.
I have a 1966 edition of the RCA Receiving Tube Manual that I bought new at Lafayette Radio Electronics in 1966 for $1.25. It still has the LRE sticker on it. Here it states that "types 3CB6, 4CB6, and 6CB6 are identical with type 6CB6A except for heater ratings". In the specs all are listed at 11 seconds except for the 6CB6.
In my 1963 RCA book it states "Type 6CB6A has a controlled heater warm-up time for use in television receivers employing series-connected heater strings."
OK, that covers it well. Either type will be OK in the red board.
I never thought about any differences between 6CB6 & 6CB6A when I built my DCPP. I bought all the tubes from Stan (ERSC), and the A's are what he had in plentiful supply. My red-board has been playing for over a year now, flawlessly, with those 6CB6A's.
Sometimes it makes a difference, and sometimes it doesn't. Tube types received a new letter for any change whatsoever. Different glass type -- new letter. Change the profile of the envelope -- new letter. Change electronic properties -- new letter.
There is a big difference between 6BQ6GTAs and 6BQ6GTBs, and it's obvious. The former has thinner cathodes, and will not stand up to spec busting the way the 'GTBs will. (In one design, the 'GTBs worked just fine where the GTAs went red plate.) As B&W TV CRTs got bigger, the HD finals had to handle mowatts, so these types got the necessary upgrades to make 'em stand up in the newer, bigger TVs.
Or 6SN7s: the originals had Pd= 3.5W, no Vgk positive operation mentioned in the specs. For the 6SN7GTB: Pd= 5.0W, Vgk positive specs provided, as are the control grid radiator "wings". (Same thing with vertical deflection types: bigger screens, mowatts.)
Something you need to look out for: was the change a minor one, like a heater characteristic made for series stringing, or was it something else?
The 'A' types are going to be much more common since so many of these were required by TV set manufacturers.
I agree that the leter suffix uprev can be for just about anything. It was supposed to be agreed apon by the RETMA and JEDEC (now part of the EIA) ald all manufacturers were to agree to the change. We all know that that is not always the case.
I keep a bunch of old tube manuals around dating back to the late 50's. I looked through them all in reference to the 6CB6A and then quoted the important info above before making any statements. In this case every book I consulted either says nothing, or clearly states something like "types 3CB6, 4CB6, and 6CB6 are identical with type 6CB6A except for heater ratings".
I've got a bunch of 6DK6 here. According to the data sheet they have a transconductance of 9800µ at 12mA. So they have even more gain than the 6CB6.
I still need to tackle transformers and put in the order for the PC board components. I hope to have something up and running in a while, but my audio budget was depleated since I bought a turntable recently.
I still need to tackle transformers and put in the order for the PC board components. I hope to have something up and running in a while, but my audio budget was depleated since I bought a turntable recently.
Would it be possible tot have about -15 volt under the 10M45 CCS in stead of -60 volts to avoid heat problems ? ( i am not using the PCB)
When breadboarding many of my LTP based experiments I routinely use a power supply that only goes to -24 volts for the CCS tail supply. I know that works good. I have seen increased distortion when using a 12 volt supply, but that was with a 6SN7 LTP. My guess is that 15 volts will be OK but I haven't really tried it. I am running 60 volts on my red boards with no heat sink. The chip gets pretty hot, but doesn't fry. It should be safe with even a tiny heat sink.
Hmm this is going to require some engineering. My PIO caps came, but they are 4" long and weigh 6oz, they are the same size as a el34. Im either going to have to mount upright or get creative on wiring the leads.