Page 696 of the RC-30 that is the circuit I was trying to find. 6CB6's driving 7027A's. I have several older RCA tube manuals at home, but I am at work now.
About 10 years ago I found a home made stereo amplifier at a local flea market. The vendor wanted $100 so I proceeded to haggle when another flea market vendor came up and told me that it didn't work and I shouldn't buy it. The two obviously knew each other and began to argue. I finally got both men to agree that sound came from both speakers but it was bad sound. The new price was now $50 which I hapilly paid. As far as I was concerned I just bought a nice pair of UTC LS-57 OPT's and the rest was a bonus.
The amp used a circuit similar to this one, but some liberties were taken with the tube line up and the component values. There might have been a similar circuit in an older tube manual, since this amp seemed older than the RC-30 (1975). It used 6L6GC's 6CB6's and a 6U8 (I think) in the amp and a 6EM7 an 0A2 and a 5U4 in the power supply. It oscillated big time, and I doubt that it ever worked. I kept the OPT's the Stancor power transformer and choke, and the tubes, tossed the rest.
Yeah, compare the old Fender guitar amps to the old RCA tube manuals.
I have collected a few bench supplies over the years including the monster 0-600 volt 0-1.5 amp HP that I use on this amp. Before I had these I used a variac feeding a 480 to 120 volt industrial transformer hooked up backwards(Antek didn't exist then). With a bridge rectifier and a big cap I had a 0-550 volt power supply. It was sufficient until I got something better.
I was thinking of using the variac that way until I get around to building a regulator around the Antek. Probably use a circuit like the IP-32, similar to what Pete did.
Did you get that Harrison beast from me?
I have a Lambda "beast" with a variac to set the voltage. I rarely use it, however, as the Heath unit sits nicely on the bench.
I got it from a friend who is in the surplus equipment business. He often gets a load of test equipment that may or may not be working. I often help him evaluate the stuff to determine how best to sell it, usually in exchange for something that I want. In this case there were several power supplies that had signs of previous water intrusion, and a lot of RF equipment. I tested and sorted everything, and kept this power supply. There are some rust stains but it works good. It cost me a day's worth of work and a road trip to Orlando, which always costs me money because I stop at ESRC.
Not sure that this beast is a Harrison unit though. I thought all the Harrisons were (overly complicated) linear supplies. This is one of the early 60 Hz switchers so it weighs about 50 pounds.
I would be tempted to make an offer, say $75 if I needed another one. There are a few things you need to know though.
In the beginning all lab power supplies were linear. They involved a high voltage supply and series pass devices to regulate the voltage down to the desired level. The usual Heathkit - Eico - Knight - etc used a high DC voltage (raw B+) in the 460 to 500 volt range. If you wanted 200 volts at 100 mA the pass device would have to dissipate 30 watts. This is wasteful and limits the output to 100 mA or so. My 200 mA Knight uses 4 6L6GC's for pass devices. The bigger supplies have a multi tapped power transformer so that the raw B+ voltage can be selected to keep the dissipation to more managable levels. The Fluke 407D is an example of this.
The second generation of HV power supplies uses a circuit that basically works like a light dimmer (a solid state variac) on the input of a big transformer - bridge rectifier - big capacitor type power supply. Some type of control loop tries to keep the output voltage or current constant. The "light dimmer" is an SCR that switches on and off at a 60Hz rate or a triac that switches at a 120 Hz rate. The low update rate combined with the sub Hz bandwidth of the CLC output filter gurantees a sluggish response and voltage overshoot.
You can use a power supply like this on tube circuits as long as you understand that it is possible and not unlikely for the output voltage to go considerably above the setpoint under certain conditions, and the current limiter may take 1 or 2 seconds to kick in due to an overload. It was a combination of these two factors that caused an electrolytic to be blown in half when it should have just sprayed its nasty goo every where. In this case the capacitor probably shorted, blowing the 1W resistor feeding it. This removed the screen voltage causing the amp to suddenly stop drawing current. This caused the power supply voltage to shoot skyward striking an arc across the open resistor. The current limiting was lazy allowing enough current to pass to vaporize the resistor and blow a 47 uf 450 V cap in half!
The third generation HV power supply is a full on high frequency switcher. The ones designed for lab use are usually quite docile and can be expected to deliver the desired voltage and current under normal conditions. The power supplies made for electrophoresis may or may not have sufficient filtering for audio use. I got two of these off of Ebay but neither worked, so I have no information as to their suitability.
I am guessing that the supply you saw is of the second or third type. The easiest way to find out is to pick it up. If it takes two hands it is the second type.
While this behavior makes things a little temperamental for experiments, frankly it behaves closer to how a "normal" unregulated power supply in a typical tube amp would anyway.
Good info, though!
This is why I don't sweat too much about pegging the needle at 150mA or more because I usually have the output voltage pegged in these situations as well...or at least close to it. I blew up the old silicon diodes in the voltage doubler once, which was an excuse to recap the old clunker as well. The film caps in the error amp were leaky, making regulation poor.
The only real concern is overheating the transformer. It is rated for 120mA and does get pretty toasty when pushed for long periods.
Does anyone know what the maximum cathode current for a 6L6 is? I haven't found it stated in any of the datasheets I have.
I think one of the best deals going is the 750-1000 Watt 1U EMS lab power supply. They are the switcher type, easily convertable from 120-240 (never mind the nameplate) and go up to 600V * 1.2A.
Here is a low voltage model for an example (no 600V ones up right now). I have one of these I use for powering hungry DHT transmitter tube filaments.
LAMBDA Power Supply md: EMS 13-75-1-D PN: 004731125 NR - eBay (item 270488848527 end time Mar-21-10 15:16:39 PDT)
They do have a noisy fan but the cooling is designed for dense, hot rack conditions and is way overkill for tube amp lab use when you're only drawing a few hundred watts max so I replace the AC blower with a DC unit (available from Allied) that runs at a much lower speed and is plenty quiet for amp lab use.
Michael
Here is another Schadey circuit: ARC ST-70 mod
http://www.arcdb.ws/ST70C3/ARC_ST70C3_schematic.gif
http://www.arcdb.ws/ST70C3/ARC_ST70C3_schematic.gif
Hi Michael,
I've heard that tempering full-on AC fan speed with a cap results in very quiet fans -- quieter than a reduced voltage DC fan. But then again, the 12vdc computer chassis fans they sell nowadays are REALLY quiet, so there are plenty of solutions available.
..Todd
Last edited:
Here is a thread on that RC-19 Schaded amplifier. Check out the SP-20 here:
http://www.diyaudio.com/forums/tubes-valves/149639-6v6-p-p-amp-rca-manual-rc-19-a-2.html
Needs some upgrading apparently.
http://www.diyaudio.com/forums/tubes-valves/149639-6v6-p-p-amp-rca-manual-rc-19-a-2.html
Needs some upgrading apparently.
I remember that thread now that I have reread it. I believe that this is the schematic from which the flea market amplifier was originally built. My memory might be foggy about exactly which tubes were in the amp when I got it, but it definitely came with metal 6L6's and a 5U4 since they are still in the box with the other parts that I kept from that amp. It is possible that the amp came with 6AU6's and I could have put 6CB6's in. Either way, when I got it the distortion was about 50% and the amp motorboated badly, so I gutted it. Did it ever work? I don't know since it looked 30 to 40 years old the caps could have all been dead. $50 for the iron was a deal.
It is obvious from the thread referenced above that there are strong opinions as to whether the circuit sucks or is pure magic. I will reserve my opinion until I have actually built and tested something along these lines. One of the posters said that an LTP phase splitter was the proper choice. I have always preferred the LTP, and more recently two cascaded LTP's. There were a few combination triode - pentode tube numbers mentioned in that thread. Tomorrow I should have a box full of $1 tubes. Some of those numbers were included in that order (by coincidence).