Miles Prower mentioned the 6X8 pentode/triode in another thread.
I wonder if something could be made of this "oddball" type in a phono preamp. I see 2 ways around the common cathode connection: red LED bias and grid leak bias. I'm partial to grid leak biased 2nd gain blocks, as they yield better bass extension. Grid leak bias in the 1st gain block requires cap. coupling between the 47 KOhm cart. loading resistor and the pentode's g1. Given a 20 MOhm grid leak value, a transparent PTFE cap. will be affordable. Perhaps 1 of the time constants can be obtained by parallelling the 20 MOhm part with capacitance.
The $64 question is whether the noise factor of the pentode is adequate.
Comments please.
I wonder if something could be made of this "oddball" type in a phono preamp. I see 2 ways around the common cathode connection: red LED bias and grid leak bias. I'm partial to grid leak biased 2nd gain blocks, as they yield better bass extension. Grid leak bias in the 1st gain block requires cap. coupling between the 47 KOhm cart. loading resistor and the pentode's g1. Given a 20 MOhm grid leak value, a transparent PTFE cap. will be affordable. Perhaps 1 of the time constants can be obtained by parallelling the 20 MOhm part with capacitance.
The $64 question is whether the noise factor of the pentode is adequate.
Comments please.
I use these for voltage regulator error amps. You can make an unbalanced differential with the triode section as the noninverting input that receives the reference voltage. The pentode half becomes the inverting input that provides the gain. Connect its control grid to the voltage set/sense divider.
I was using 6KE8s for this, but these guys are a good deal more expensive, and 6X8s are quite common, nobody wants them, and I can get 'em dirt cheap.
I was using 6KE8s for this, but these guys are a good deal more expensive, and 6X8s are quite common, nobody wants them, and I can get 'em dirt cheap.
Eli Duttman said:
Since this type was intended for use as a cathode coupled mixer/oscillator, it looks like the pentode section would be quiet enough for a phono preamp. It was designed for use in a typically low level part of the receiver. It also looks to have some very low internal capacitances as well. The triode section looks to have good linearity as well. You can always get around that common cathode by applying fixed bias, or adding a bit of positive grid bias to offset the cathode bias.
Noise and microphonic effect are quite typical troubles with non-specialized tubes in this application.
If you don't need to buy anything except such cheap tubes I think the best response is trying all the possible solutions you mentioned.
Other possible cheap candidates you might consider are the 20EZ7 (basically a low noise and low cost 12AX7 with 20V filament supply and different pin-out) and the 5755.
IMO, the 5755 as input tube of a MM phono stage can really stand out.
120V anode voltage and 0.5 mA could be a good working point (the bias is -0.5V).
In my experience the best phono amps all had a passive split RIAA, with the 75 us network after the second stage. With some expedients (basically a design choice), such network can have a negligible insertion loss.
45
If you don't need to buy anything except such cheap tubes I think the best response is trying all the possible solutions you mentioned.
Other possible cheap candidates you might consider are the 20EZ7 (basically a low noise and low cost 12AX7 with 20V filament supply and different pin-out) and the 5755.
IMO, the 5755 as input tube of a MM phono stage can really stand out.
120V anode voltage and 0.5 mA could be a good working point (the bias is -0.5V).
In my experience the best phono amps all had a passive split RIAA, with the 75 us network after the second stage. With some expedients (basically a design choice), such network can have a negligible insertion loss.
45
45 said:
Not necessarily. You never know until you try. The spec sheet for 6BQ7s make no mention of any sort of audio useage at all. Yet, when used to make a cascoded LTP with active tail load, these really rawk.
An externally hosted image should be here but it was not working when we last tested it.
More gain than the usual solutions (12AX7, 6SL7) none of that pentode nastiness problem mentioned with pentode LTPs. These, too, are cheap since they have no audiophool pedigree. I also discovered that the Sylvania and Motorola versions with the parallel connected heaters aren't microphonic at all. (Different story with the ones with series connected heaters.)Besides, the fat cathodes score extra glowey bottle coolness points.
6BQ6GTBs also are "non-specialized" with no mention than anything other than TV horizontal deflection duty. Sounds as good as PP 6V6s, but with better than twice the output power. Given the need for pulling bigamps through the deflection coils, these require less impedance transformation, so that the OPTs are easier to design, and work better. Also, cheap with no audiophool pedigree. (But they've been going up lately. I guess the good folks at the VT store figured something was up when I bought a whole bunch at $1.30 a pop.)
Triode-strapping the pentode section may be an idea.
According to Tim,
"After some testing, I discovered that it's (6X8) really quite reasonable, its apparent similarity to the 6J6 being no accident. In fact, with the pentode section wired as triode, it may well be identical to a 6J6."
From:
http://webpages.charter.net/dawill/tmoranwms/Elec_Compound2.html
Just a pre-coffee thought.
According to Tim,
"After some testing, I discovered that it's (6X8) really quite reasonable, its apparent similarity to the 6J6 being no accident. In fact, with the pentode section wired as triode, it may well be identical to a 6J6."
From:
http://webpages.charter.net/dawill/tmoranwms/Elec_Compound2.html
Just a pre-coffee thought.
It works OK with really low noise floor sans shield cap and zero microphonics even when knocked on with finger nail. I get 60dB from cart in to line out using a Lundhal 9206 set at 10X, this Ruskie, passive RIAA, and 6N8. 2 Mu followers like in the Dude. No weird interactions. Riaa curve is the same with or without the step up.
Will show that phono here soon.
Will show that phono here soon.
Speaking of the 6J6, I've thought of using one in a passive equalization phono preamp using LED bias at the common cathode, figuring that the low dynamic impedance would more or less decouple the stages from each other. Some more bias current for the LED via a separate JFET or ring-of-two current source would lower the LED impedance some more and help out the isolation between stages. I've got a fair number of 6J6s lying about, so I figure I might as well do something with them. This would probably work with the 6X8, too, though I only have a couple of those.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.