I don't know the 6P43P-E at all. But I suspect your disappointing experiences with it might result from it's different properties in comparison with the 6CW5/EL36/6P31S. The circuitry might be in need of some tweaking.
Best regards!
In triode mode the mu and DC conditions line up between the 6P43P-E and 6CW5 almost perfectly. Pentode mode they do seem to differ enough that some extra care is needed.
I’ve heard from some that they do not have stable bias on higher plate voltages when pentode wired, might be part of the issue in my particular amplifier I tried them in.
Not disappointing at all, just means additional work is needed to use them in pentode. As I have twenty-one 6P43P-E on hand and was planning them mostly for triode mode projects I’m in no huge hurry.
I’ve heard from some that they do not have stable bias on higher plate voltages when pentode wired, might be part of the issue in my particular amplifier I tried them in.
Not disappointing at all, just means additional work is needed to use them in pentode. As I have twenty-one 6P43P-E on hand and was planning them mostly for triode mode projects I’m in no huge hurry.
I just might. I have some toroids on hand that will come out to 3k~ or so with an 8 ohm load, should be good for a single or parallel pair in triode- minor differences (1.5~2 watts) in power output between 6k and 3k reflected load. Might just use them for the V8 build, 4 per side, 6N1P drivers, since I already have a couple sockets wired up for the volt-amp/concertina front end. Just gotta buy a new hole saw as I smoked my last one that I usually use for noval sockets.
Only problem is my power supply toroid that's set aside will give way more volts than I need at 2x150VAC, unless I parallel them to run lower volts and higher current. Then I would need to look at a different driver maybe? You think a 6N1P concertina can swing the volts with only a 200 volt supply? Assuming fixed bias, Power per 6P43P-E pair at 200 volts will drop to like ~5 watts out with 6k, 7 watts with 3k, which seems wasteful honestly. Drive requirements are cut almost in half though.
Maybe garter bias with a 375 volt supply, drop 25-30 volts in the PSU, chew up some of the extra volts in the cathodes
It would only be four 1k, 3-5 watt resistors per pair. 16 resistors total. Not to0 bad, only double the count for regular cathode bias. Not worried about waste heat, it's for the workshop that's always cold anyway.
That would net maybe ~12-13 watts out of a pair even with a 6k reflected load of my 3k outputs, which means two pairs would be ~25 watts. Pushing it for a 25VA toroid though. I also have a pair of 50VA that will be a reflectd 4k load, which would lose me a watt per pair looking at 8k per pair. Still not too bad at ~22 watts out.
Hmmm.
Money is tight so I'm not going to be spending more money on transformers anytime soon, trying to build with what's on hand, y'know?
Garter bias sounds fun if they will be stable at that voltage.
Only problem is my power supply toroid that's set aside will give way more volts than I need at 2x150VAC, unless I parallel them to run lower volts and higher current. Then I would need to look at a different driver maybe? You think a 6N1P concertina can swing the volts with only a 200 volt supply? Assuming fixed bias, Power per 6P43P-E pair at 200 volts will drop to like ~5 watts out with 6k, 7 watts with 3k, which seems wasteful honestly. Drive requirements are cut almost in half though.
Maybe garter bias with a 375 volt supply, drop 25-30 volts in the PSU, chew up some of the extra volts in the cathodes
It would only be four 1k, 3-5 watt resistors per pair. 16 resistors total. Not to0 bad, only double the count for regular cathode bias. Not worried about waste heat, it's for the workshop that's always cold anyway.That would net maybe ~12-13 watts out of a pair even with a 6k reflected load of my 3k outputs, which means two pairs would be ~25 watts. Pushing it for a 25VA toroid though. I also have a pair of 50VA that will be a reflectd 4k load, which would lose me a watt per pair looking at 8k per pair. Still not too bad at ~22 watts out.
Hmmm.
Money is tight so I'm not going to be spending more money on transformers anytime soon, trying to build with what's on hand, y'know?
Garter bias sounds fun if they will be stable at that voltage.
Last edited:
Low supply voltage, high signal voltage demand and Concertina won't fit well. You'd better look for another PI design, maybe a LTP or a Paraphase.
Best regards!
Admittedly, I'm not much of an adept of the Concertina, mostly due to it's very different output impedances and it's relatively limited output voltage swing. Instead, I prefer the LTP with a SS CCS tail, especially if the power tubes are fix biased and a negative voltage rail is available.
Best regards!
Best regards!
More voltage on the front end than the power tubes…. But you only get half or so the VA rating out of the power trafo with half the winding seeing almost all the load.
18 volts output uses up all the headroom in that concertina. 64 volts at 3.8mA is damn near vg=0. More power hungry tubes and you’d need a higher supply. Not that I haven’t abused them myself or anything - my latest cute little amplifier runs the concertina within an inch of its life and still sounds clean… with no NFB at all.
18 volts output uses up all the headroom in that concertina. 64 volts at 3.8mA is damn near vg=0. More power hungry tubes and you’d need a higher supply. Not that I haven’t abused them myself or anything - my latest cute little amplifier runs the concertina within an inch of its life and still sounds clean… with no NFB at all.
PL802 falls in the 12HG7/BY7/ rank....cheap, fat cathode, with high gm. In audio work I use them as triodes giving excellent 25-28dB in RC config. There are a couple of caveats using these tube types.
1. Note high g1-Cath capacitance on data sheet. Use no greater than 100R as a g1 stopper. If one goes to 10K or sim, then HF response falls off fast. Using a low value grid leak pushes the onus on to the previous stage that has a lowish output impedance.. Otherwise one could be blaming another component for lost HF response!
2. Through the 1950´s through to early 1970´s, these tube variants were produced in the millions, and consistency especially H2.H5 distortion varies enormously.
To get sensible matched pairs with equal Va, distortion etc I needed to delve through my stock of several hundred NOS to achieve 50 matched pairs. Tedious work. Some were way, way out, but who cared for video distortion when the specs in those early days were very lax ?
BB
1. Note high g1-Cath capacitance on data sheet. Use no greater than 100R as a g1 stopper. If one goes to 10K or sim, then HF response falls off fast. Using a low value grid leak pushes the onus on to the previous stage that has a lowish output impedance.. Otherwise one could be blaming another component for lost HF response!
2. Through the 1950´s through to early 1970´s, these tube variants were produced in the millions, and consistency especially H2.H5 distortion varies enormously.
To get sensible matched pairs with equal Va, distortion etc I needed to delve through my stock of several hundred NOS to achieve 50 matched pairs. Tedious work. Some were way, way out, but who cared for video distortion when the specs in those early days were very lax ?
BB
That high gm of video pentodes lends them to application as k-followers with some kick to them. Cheaper and more power efficient than “wasting” EL84’s or 6V6’s in the application. G-K capacitance almost irrelevant - just like when using a mosfet.
Also attractive in using pairs for small guitar amps. One doesn't need need to worry about capacitance or consistency, if you’ve put in a way to trim bias and at least null out the DC. The high mu triode/video pentode bottles can give you a lot of gain.
Also attractive in using pairs for small guitar amps. One doesn't need need to worry about capacitance or consistency, if you’ve put in a way to trim bias and at least null out the DC. The high mu triode/video pentode bottles can give you a lot of gain.
I have a theory that a lot of TV type tubes were manufactured for a specific purpose, probably as part of a contract to supply an equipment manufacturer, and hence only had to meet the criteria to fulfil that role, whereas the 6.3V heater variants were usually manufactured for different roles, and had a greater need to meet the minimum quality requirements of the tube data sheet.
When they were manufacturing these tubes in their different variants, does anyone know when they decided to use the same tube construction with the different heater? Could there have been the possibility that 'B'-grade innards ended up in serial heater TV type roles?
I don't doubt your statement, but am astonished anyway. In the 1970ies PL802 had become an extremely scarce, sought after, and expensive tube. Tube manufacturers were even forced to develop and market SS replacements like this one by Valvo/Philips (sorry, in German).
Indeed TV tubes were specialized gradually to fulfill their tasks. I remember an old European (by Philips iirc) TV chassis that featured four ECL80's in four different roles: AF amplifier, sync separator, sweep oscillator, frame deflection. Or think of the PCL82 that originally was used as AF amplifier and for frame deflection, before separation occured into the PCL86 and PCL85/805 respectively.
But I have no indication that TV tubes were just B grade. Quite the opposite, I'd say, as many TV sets were run as long as program was broadcasted, and manufacturers really had no interest to see their reputation get ruined.
Best regards!