At some point I've bought a couple of ECH83 and EBF83 tubes and now I'm thinking about how to use them. They're european parallel to american space-charge tubes: low voltage anodes.
ECH83 is a triode-heptode with 6.3V/0.3A filament and 6-50V plates.
ECH83 datasheet
I've seen the heptode might be also used for rf/if amplification, so I've tried to think of audio amplification (from my experience most of RF tubes may happily amplify audio). I've never made a heptode audio stage so this is a funny experiment for me. Here's an attempt to design something -
whatcha think? will this thing smoke?
ECH83 is a triode-heptode with 6.3V/0.3A filament and 6-50V plates.
ECH83 datasheet
I've seen the heptode might be also used for rf/if amplification, so I've tried to think of audio amplification (from my experience most of RF tubes may happily amplify audio). I've never made a heptode audio stage so this is a funny experiment for me. Here's an attempt to design something -
whatcha think? will this thing smoke?
Hello!
No, it won't smoke certainly ;-), and most certainly it will work, too. But think of noise level.... The more grids in a tube, the higher the noise level.... hexodes, heptodes and pentagrid converters are quite noisy. Also, microphony might be an issue with all these grids. So you were right putting the triode before the heptode.
Many years ago I built a similar amp with an AK2 - a pre-war octode with 4 Volts heating, just for fun. Impressive bulb, quite noisy and incredibly microphonic... (the tube was in the first stage...) Even got acoustic feedback from the speaker.... But - it amplified.... ;-) But with those modern types, you shouldn't have these problems almost anymore.
I have two questions:
1. biasing on common cathode - are the two simple grid leak resistors sufficient? Especially the 47 k on the triode grid most certainly will be much too small - I would suggest 1 Mohm at least. Its low value comes from its operation as an oscillator where grid bias was generated by rectification of the rf input signal on the grid-cathode diode.
2. Why do you use the second regulator grid of the heptode as a (3rd) screen grid? Why not using it in parallel to the first grid as a second regulator grid, as it usually has been intended to? Thus enhancing sensitivity? Would perhaps be worth trying.
Let us know about your results! I also have a lot of these tubes lying around....
Good success!
Uli
No, it won't smoke certainly ;-), and most certainly it will work, too. But think of noise level.... The more grids in a tube, the higher the noise level.... hexodes, heptodes and pentagrid converters are quite noisy. Also, microphony might be an issue with all these grids. So you were right putting the triode before the heptode.
Many years ago I built a similar amp with an AK2 - a pre-war octode with 4 Volts heating, just for fun. Impressive bulb, quite noisy and incredibly microphonic... (the tube was in the first stage...) Even got acoustic feedback from the speaker.... But - it amplified.... ;-) But with those modern types, you shouldn't have these problems almost anymore.
I have two questions:
1. biasing on common cathode - are the two simple grid leak resistors sufficient? Especially the 47 k on the triode grid most certainly will be much too small - I would suggest 1 Mohm at least. Its low value comes from its operation as an oscillator where grid bias was generated by rectification of the rf input signal on the grid-cathode diode.
2. Why do you use the second regulator grid of the heptode as a (3rd) screen grid? Why not using it in parallel to the first grid as a second regulator grid, as it usually has been intended to? Thus enhancing sensitivity? Would perhaps be worth trying.
Let us know about your results! I also have a lot of these tubes lying around....
Good success!
Uli
I can't say if it will work, but that looks interesting as I have a few "unusual" tubes myself that I'd like to use. Quite a few ECH41 (triode-hexode) and UAF42 (diode-pentode), and even if only one tube section is usable it would be fun building something with them.
And if someone knows if something could be built using an EF9 pentode, please let me know
And if someone knows if something could be built using an EF9 pentode, please let me know
showdown said:And if someone knows if something could be built using an EF9 pentode, please let me know
Hi!
The EF 9 is the almost identical type to the EF 89, only with different bulb and base, it is a remote-cutoff-pentode with gliding screen voltage (reduction of distortion) for variable rf or if amplification in radio sets. But you also can use it for af amplification. If I remember well I even have seen a music dynamics compressor/enhancer which was using the variable amplifiaction (depending on bias voltage) of this tube. I think it could be used similarly to the EF 83, which is the remote-cutoff-version of the EF 86, both types specially crafted (low hum sensitivity) for af amplification service.
I would try the EF 9 in similar layouts to the EF 86, just playing around with cathode and screen resistor values - both certainly smaller than with the low-current EF 86. I would suggest 100 k plate, 330 k screen and 270 Ohms cathode - just to start with. Grid leak 1 MOhm. And also trying different grid bias settings to obtain different amplification factors (cathode on ground and bias fed from a variable voltage source).
Triode operation also should possible.
Good success!
Uli
hmm
2 ulibub:
thanks for the reply!!!!
yes, 47K grid leak resistor looks small. Still the datasheet specifies it as the bias resistor and I think it's reasonable to try it like this first. If the signal comes from something between 600 Ohm (microphone) to 8K (pickup) it should work (actually the triode will work in grid leak mode with 600 to 8000 ohm resistor).
now I don't know a thing about grids in heptode! The datasheet says that if heptode is operated as rf/if amplifier, Vg2+Vg3+Vg4 = 6 to 25 V. I thought that means they're all connected to the B+.
If I connect the control grid to the g3 there's going to be a positive grid (g2) between the g1 and g3 - and how the hell this thing should speed up electrons on their way to the plate? Backwards?
I think I'm going to read some more about heptodes.
The way it's wired in the schem it acts like a pentode.
The questions I have are more about the resistor between the stages (100K) - I'm not sure I can connect the triode's anode to the grid of the heptode without a resistor because of the common cathode. If I could eliminate it I would certainly get a louder chain - if the heptode won't distort too much.
Also the load resistors are a question - the little schem in the datasheet shows the heptode with the g2 and g4 wired to B+ and the plate a lower potential. I've put a 150K load on the "screens" and 100K on the plate to make the plate more positive. Am I wrong? Should I wire the screens straight to B+ ?
2 ulibub:
thanks for the reply!!!!
yes, 47K grid leak resistor looks small. Still the datasheet specifies it as the bias resistor and I think it's reasonable to try it like this first. If the signal comes from something between 600 Ohm (microphone) to 8K (pickup) it should work (actually the triode will work in grid leak mode with 600 to 8000 ohm resistor).
now I don't know a thing about grids in heptode! The datasheet says that if heptode is operated as rf/if amplifier, Vg2+Vg3+Vg4 = 6 to 25 V. I thought that means they're all connected to the B+.
If I connect the control grid to the g3 there's going to be a positive grid (g2) between the g1 and g3 - and how the hell this thing should speed up electrons on their way to the plate? Backwards?
I think I'm going to read some more about heptodes.
The way it's wired in the schem it acts like a pentode.
The questions I have are more about the resistor between the stages (100K) - I'm not sure I can connect the triode's anode to the grid of the heptode without a resistor because of the common cathode. If I could eliminate it I would certainly get a louder chain - if the heptode won't distort too much.
Also the load resistors are a question - the little schem in the datasheet shows the heptode with the g2 and g4 wired to B+ and the plate a lower potential. I've put a 150K load on the "screens" and 100K on the plate to make the plate more positive. Am I wrong? Should I wire the screens straight to B+ ?
Simply try it!!!
It certainly will work, the question is HOW GOOD it will work! And instead of much discussion, just build it and see the results! Thats one of the big advantages of tube technics against semiconductors... It still works, even if it might be far from optimal. Don't know how triode operation is behaving with the remote-cutoff characteristics. It depends largely on the gliding screen voltage to reduce distortion - these tubes otherwise have quite some nonlinear characteristic. So I would prefer pentode operation.
Or, use "pseudo-triode" service without condenser on the screen grid, only fed by its resistor to keep the gliding voltage characteristic. So you should get quite low-distortion triode operation.
But - simply try it and find the best operation mode!
Uli
It certainly will work, the question is HOW GOOD it will work! And instead of much discussion, just build it and see the results! Thats one of the big advantages of tube technics against semiconductors... It still works, even if it might be far from optimal. Don't know how triode operation is behaving with the remote-cutoff characteristics. It depends largely on the gliding screen voltage to reduce distortion - these tubes otherwise have quite some nonlinear characteristic. So I would prefer pentode operation.
Or, use "pseudo-triode" service without condenser on the screen grid, only fed by its resistor to keep the gliding voltage characteristic. So you should get quite low-distortion triode operation.
But - simply try it and find the best operation mode!
Uli
Re: hmm
Hi Engels!
))
Yes, the first screen grid (g2) - between the two regulator grids (g1 + g3) - protects g1 from capacitive couplings of the oscillator frequency, which is fed into the mixer via g3, to avoid oscillator frequency being radiated away by the antenna, which is fed into g1. It also accelerates the electrons to be fast enough to penetrate g3 (as in the old space-charge tubes), so it forms some kind of "virtual cathode", an electron cloud directly in front of g3. The rest of the system (second screen, g4, and suppressor, g5) acts as a simple pentode. That's the usual mixer operation of these tubes.
But with all these grids you also can configure special conductivity conditions depending on the wave angles of the different ac voltages on the different grids relative to another. - the "Enneode" (type EQ80)- with 3 regulator grids, is such a type and was used as an amplifying FM demodulator. In this way, ECH83 often were used as amplitude filters in old TV sets to get the syncronisation pulses out of the video signal.
Simply take it away (ot put the grid leak in front of it, to avoid the voltage divider) and take an ordinary grid stopper directly in front of the grid - 1 to 10 k should be sufficient.
I only can tell from the ECH81 or ECH41/42 in radio sets, where the screen voltage always has to be much lower than the plate. I don't know how it is with these low-voltage tubes. Just try, what sounds best...
Uli
Hi Engels!
engels said:
))Yes, the first screen grid (g2) - between the two regulator grids (g1 + g3) - protects g1 from capacitive couplings of the oscillator frequency, which is fed into the mixer via g3, to avoid oscillator frequency being radiated away by the antenna, which is fed into g1. It also accelerates the electrons to be fast enough to penetrate g3 (as in the old space-charge tubes), so it forms some kind of "virtual cathode", an electron cloud directly in front of g3. The rest of the system (second screen, g4, and suppressor, g5) acts as a simple pentode. That's the usual mixer operation of these tubes.
But with all these grids you also can configure special conductivity conditions depending on the wave angles of the different ac voltages on the different grids relative to another. - the "Enneode" (type EQ80)- with 3 regulator grids, is such a type and was used as an amplifying FM demodulator. In this way, ECH83 often were used as amplitude filters in old TV sets to get the syncronisation pulses out of the video signal.
Simply take it away (ot put the grid leak in front of it, to avoid the voltage divider) and take an ordinary grid stopper directly in front of the grid - 1 to 10 k should be sufficient.
I only can tell from the ECH81 or ECH41/42 in radio sets, where the screen voltage always has to be much lower than the plate. I don't know how it is with these low-voltage tubes. Just try, what sounds best...
Uli
Re: Re: hmm
thanks a big lot!
I think I will make a switch for the third grid to connect it to the screens or to the control grid.
Moving the grid bias resistor before the grid stopper seems 100% right.
As for the voltages in low voltage tubes - I've got an amplifier with 12DL8 which is a space-charge grid tetrode. The space charge grid is placed between the control grid and the cathode (not between the grid and the plate as in a "normal" tetrode). The space charge grid is wired directly to 12V and the plate has something like 6V on it. It works fine. The ECH83 heptode is not a space-charge tube so it doesn't need that potential on the screen but who knows... those tubes always work - even if you put 100V on screens and 1V on plates... they're tubes.
ulibub said:
thanks a big lot!
I think I will make a switch for the third grid to connect it to the screens or to the control grid.
Moving the grid bias resistor before the grid stopper seems 100% right.
As for the voltages in low voltage tubes - I've got an amplifier with 12DL8 which is a space-charge grid tetrode. The space charge grid is placed between the control grid and the cathode (not between the grid and the plate as in a "normal" tetrode). The space charge grid is wired directly to 12V and the plate has something like 6V on it. It works fine. The ECH83 heptode is not a space-charge tube so it doesn't need that potential on the screen but who knows... those tubes always work - even if you put 100V on screens and 1V on plates... they're tubes.
Hi engels!
Nice to see someone is working with space-charge tubes. I don't remember ever having seen these tubes in use troughout Europe. So I know them only from theory. How are they performing? Do you have a circuit diagram of that amp?
And yes, low-voltage tubes doesn't mean they're built exclusively for battery operation, they work equally as fine with "normal" tube voltages. They're only designed still to function at low voltages, too. In my phono preamp I use an ECC86, also a low-voltage type, in SRPP with 50 Volts on each triode. Works perfectly.
Greetings, Uli
Nice to see someone is working with space-charge tubes. I don't remember ever having seen these tubes in use troughout Europe. So I know them only from theory. How are they performing? Do you have a circuit diagram of that amp?
And yes, low-voltage tubes doesn't mean they're built exclusively for battery operation, they work equally as fine with "normal" tube voltages. They're only designed still to function at low voltages, too. In my phono preamp I use an ECC86, also a low-voltage type, in SRPP with 50 Volts on each triode. Works perfectly.
Greetings, Uli
space charge
yep, I'm addicted to weird tubes. Space charge tubes are pretty cool - cheap and the iron is even cheaper. The only iron you can freely buy today at almost any electronics shop - 12V power (for fluorescent lamps) and line output trannies.
If you're interested, here's the amp I've made -
Space Charger amplifier - schematics in PDF format
It's the greatest thing I've done. I play guitar and bass and at the moment this is the only amp I record with. Very quiet but sounds huge.
I would not call this a hi-fi amplifier. The preamp tubes preform well but for some reason they don't sound too hi-fi. I've listened to the preamp only and it cuts some of the high frequencies. It might as well be the capacitors in the chain, I'm not sure. The resulting sound is very pleasing so I've left it like that. It's more like a germanium transistor that gives everything a nice vintage flavour.
Now I'm thinking about making a spacecharge amp with PP output... but the PI is a big question.
yep, I'm addicted to weird tubes. Space charge tubes are pretty cool - cheap and the iron is even cheaper. The only iron you can freely buy today at almost any electronics shop - 12V power (for fluorescent lamps) and line output trannies.
If you're interested, here's the amp I've made -
Space Charger amplifier - schematics in PDF format
It's the greatest thing I've done. I play guitar and bass and at the moment this is the only amp I record with. Very quiet but sounds huge.
I would not call this a hi-fi amplifier. The preamp tubes preform well but for some reason they don't sound too hi-fi. I've listened to the preamp only and it cuts some of the high frequencies. It might as well be the capacitors in the chain, I'm not sure. The resulting sound is very pleasing so I've left it like that. It's more like a germanium transistor that gives everything a nice vintage flavour.
Now I'm thinking about making a spacecharge amp with PP output... but the PI is a big question.
Tubes at 12V +B...
Ive recently been fooling with various 'Standard' triodes at 12-20V +B. Take a look at the scheme I posted around halfway down this page....
http://www.diyaudio.com/forums/showthread.php?s=&threadid=89195&perpage=10&pagenumber=5
Ive tried 6SN7, 5687, ECC88/6DJ8, and 7586 Nuvistors. All work reasonably well, The Nuvistors particularly well, being very transparent, providing you keep the input below 1V RMS, if your supply is 13.6V, as it runs out of 'headroom'
I have not tested for distortion/noise etc, But it Does Sound pretty good, with no really apparent 'noise' or flaws-In fact, running it via my OTL,Ii have difficulty in telling whether its switched in or out of circuit, so the dist/noise figures cant be too bad.......
IF you have a 15-20V supply, the circuit performs extremely well, and does not have input limitations as on 13.6V It has a gain of around 4-5, and is happy running into 'loads' of 47K-100K.- Great little 'scrap-box' project
Ive recently been fooling with various 'Standard' triodes at 12-20V +B. Take a look at the scheme I posted around halfway down this page....
http://www.diyaudio.com/forums/showthread.php?s=&threadid=89195&perpage=10&pagenumber=5
Ive tried 6SN7, 5687, ECC88/6DJ8, and 7586 Nuvistors. All work reasonably well, The Nuvistors particularly well, being very transparent, providing you keep the input below 1V RMS, if your supply is 13.6V, as it runs out of 'headroom'
I have not tested for distortion/noise etc, But it Does Sound pretty good, with no really apparent 'noise' or flaws-In fact, running it via my OTL,Ii have difficulty in telling whether its switched in or out of circuit, so the dist/noise figures cant be too bad.......
IF you have a 15-20V supply, the circuit performs extremely well, and does not have input limitations as on 13.6V It has a gain of around 4-5, and is happy running into 'loads' of 47K-100K.
- Great little 'scrap-box' project
I would start with "pseudo-triode" mode as I described above, wired as a true pentode, screen with its own resistor and NOT connected to the plate, but WITHOUT screen grid condenser. I haven't tried it so far, but would start this way. And this should apply to all remote-cutoff-rf-pentodes, like EF 9/11/41/85/89/93/183...., just to mention the most frequent ones.
I am very interested in your results! Don't have the time at the moment to try it myself...
Uli
I am very interested in your results! Don't have the time at the moment to try it myself...
Uli
Wow - perfect! Looking to explore space charge tubes (and tubes in general) I've acquired a single ECH83.
Would there be any problem running filament and B+ from the same supply? It can do 12V at 5A, and I can use a hefty 6V regulator for the filament.
I really want to build one of these but I'm having trouble sourcing the tubes.
Would there be any problem running filament and B+ from the same supply? It can do 12V at 5A, and I can use a hefty 6V regulator for the filament.
I really want to build one of these but I'm having trouble sourcing the tubes.
Hi Cursor!
No problem running the heaters from the same supply.
But looking at the diagram on your link - I get some headache looking at the 100 kohm resistors in the cathode lines of the 12U7 tubes - if they are correct. Using about half or one third of the plate resistor voltage drop for bias alone doesn't leave much voltage across the tube - I wouldn't expect neither a high amplification nor a good overdrive reserve.
For the triodes you could also use the double-triode ECC 86. But don't know at the moment about its plate and bias resistors. Don't know about the space-charge power tube - EF 98 is a low-voltage (car-radio) pentode wich also stands some plate dissipation.
Uli
No problem running the heaters from the same supply.
But looking at the diagram on your link - I get some headache looking at the 100 kohm resistors in the cathode lines of the 12U7 tubes - if they are correct. Using about half or one third of the plate resistor voltage drop for bias alone doesn't leave much voltage across the tube - I wouldn't expect neither a high amplification nor a good overdrive reserve.
For the triodes you could also use the double-triode ECC 86. But don't know at the moment about its plate and bias resistors. Don't know about the space-charge power tube - EF 98 is a low-voltage (car-radio) pentode wich also stands some plate dissipation.
Uli
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.