Cga of 0.6pf presumably is in pentode mode. Will be much higher in triode connection and dominates the calculation result.Many thanks! Now I have to admit I didn't consider how to calculate with the multiple elements!I was only familiar with the basic triode calculation;
Cin = Cgk + Cgp*(A+1)
Irrespective, my thinking was that the EL12N appeared to be a harder tube to drive., especially if there was significant additional capacitance to drive on the input from cables etc.
Thanks Gupsta, much appreciated. Would you know how to calculate, or a guess? No problem if not!
My intent is NOT to use the EL12N by the way, just trying to understand how hard it is to drive relative to other triodes.
I have read Ale's slew rate articles and done some calculations with usual audio triode suspects.. and yes I understand the 01A doesn't have much drive.
I'm just curious about how much could be optimised as far as its implementation - reducing stray capacitance as much as possible, not using long interconnects, being conscious of the miller input capacitance of following stage etc.
My intent is NOT to use the EL12N by the way, just trying to understand how hard it is to drive relative to other triodes.
I have read Ale's slew rate articles and done some calculations with usual audio triode suspects.. and yes I understand the 01A doesn't have much drive.
I'm just curious about how much could be optimised as far as its implementation - reducing stray capacitance as much as possible, not using long interconnects, being conscious of the miller input capacitance of following stage etc.
Your welcome.
It's not something I have researched. Vaguely recall one of Ale's blog entires measuring it. A google search suggests a few things. But with such small capacitances, a simple multimeter isn't going to cut it.
I gather G2 acts as a sheild which is why you see such low figures in pentode data sheets between grid and anode, usually 0.xxx pf. But for triodes or pentodes in triode mode that sheilding effect isn't there so capacitance is much higher. The bigger the anode the higher the capacitance tends to be, but bigger anodes reduce Ra so always trade a offs, there will be other factors too. Some of the tantalum thoriated tubes boast rather low pf, but need a fair amount of voltage and may have gas issues, then there are in directly heated tubes; even12ax7 has low pf.
The tube socket will also introduce capacitance. Teflon sockets might reduce that, or maybe no socket and hollow banana plugs, might be a way of reducing it.
Ultimately (as stated by others) the solution would be the source follower.
There are people here more expert than me who may need to correct me etc. Only replied as I contributed earlier.
Best wishes
It's not something I have researched. Vaguely recall one of Ale's blog entires measuring it. A google search suggests a few things. But with such small capacitances, a simple multimeter isn't going to cut it.
I gather G2 acts as a sheild which is why you see such low figures in pentode data sheets between grid and anode, usually 0.xxx pf. But for triodes or pentodes in triode mode that sheilding effect isn't there so capacitance is much higher. The bigger the anode the higher the capacitance tends to be, but bigger anodes reduce Ra so always trade a offs, there will be other factors too. Some of the tantalum thoriated tubes boast rather low pf, but need a fair amount of voltage and may have gas issues, then there are in directly heated tubes; even12ax7 has low pf.
The tube socket will also introduce capacitance. Teflon sockets might reduce that, or maybe no socket and hollow banana plugs, might be a way of reducing it.
Ultimately (as stated by others) the solution would be the source follower.
There are people here more expert than me who may need to correct me etc. Only replied as I contributed earlier.
Best wishes
I believe I have a workable solution for a two-stage all-DHT SET with 01A driver. The solution is 4-prong:
1. 01A running at 7-8 mA current. This may sound like abusing the tube, but it is not. There is nothing in data sheet that says 3 mA is the maximum current. The Cunningham data sheet says "... as voltage amplifier, CX-301A should be used according to tabulated data AND GRAPHS. The plate graph extends to currents slightly above 7 mA. Moreover, there is article by G.M. Best, "Vacuum Tube Data" in the vol.6 (July) 1924 Radio magazine, in which plate current for UV-201A and C-301A is listed as 1.2 - 8 mA. For those who don't believe, the issue is available on Google Books.
Higher current doubles the driving capability of 01A in terms of slew rate into capacitive load. It reduces Rp from 10K to less than 7K, which improves conditions for transformer coupling (#3).
2. There are versions of 01A with even lower Rp, about 4K. I call them "Super 01A".
3. 1:2.2 step-up interstage transformer. With 2 Vrms source, 01A can output 16 Vrms, which is about the maximum for this tube that should not be exceeded. 16 Vrms is not enough to drive a power DHT. Stepping-up input signal is not a good idea, as it will exceed low distortion output capability of 01A. Although 16×2.2=35.2 Vrms may be a little short, it is in the right ballpark.
Step-up IT will degrade driver's current capability, but to counter that, there is the third prong:
3. Use power DHT with substantially less Miller capacitance than that of 2A3, 300B, or EL-12.
I am finalizing the prototype, and will open a new thread about it when I am done.
1. 01A running at 7-8 mA current. This may sound like abusing the tube, but it is not. There is nothing in data sheet that says 3 mA is the maximum current. The Cunningham data sheet says "... as voltage amplifier, CX-301A should be used according to tabulated data AND GRAPHS. The plate graph extends to currents slightly above 7 mA. Moreover, there is article by G.M. Best, "Vacuum Tube Data" in the vol.6 (July) 1924 Radio magazine, in which plate current for UV-201A and C-301A is listed as 1.2 - 8 mA. For those who don't believe, the issue is available on Google Books.
Higher current doubles the driving capability of 01A in terms of slew rate into capacitive load. It reduces Rp from 10K to less than 7K, which improves conditions for transformer coupling (#3).
2. There are versions of 01A with even lower Rp, about 4K. I call them "Super 01A".
3. 1:2.2 step-up interstage transformer. With 2 Vrms source, 01A can output 16 Vrms, which is about the maximum for this tube that should not be exceeded. 16 Vrms is not enough to drive a power DHT. Stepping-up input signal is not a good idea, as it will exceed low distortion output capability of 01A. Although 16×2.2=35.2 Vrms may be a little short, it is in the right ballpark.
Step-up IT will degrade driver's current capability, but to counter that, there is the third prong:
3. Use power DHT with substantially less Miller capacitance than that of 2A3, 300B, or EL-12.
I am finalizing the prototype, and will open a new thread about it when I am done.
Interesting concept, and I'm in total agreement about an all-DHT SE amp. As you know, I have a similar 2 stage all-DHT 2a3 amp working as my daily amp, and loving it. I've used the 2P29L as the input stage which makes a lot of things simpler and I use a 1:3 SUT on the input to get the gain. Cheaper to get a smaller input SUT than an interstage. Even a Hammond 124B sounds good as a SUT, and that's what I am using pending the arrival of a more upmarket SUT.
But yours is an interesting and original challenge. Is the 01A the right tube for this job? Set up right it's as good a tube as you can find - no argument there. The datasheets are ancient and don't specify maximum plate dissipation, so as you say it's mostly guesswork. If you can play around with this and make it work it would be very informative!
But yours is an interesting and original challenge. Is the 01A the right tube for this job? Set up right it's as good a tube as you can find - no argument there. The datasheets are ancient and don't specify maximum plate dissipation, so as you say it's mostly guesswork. If you can play around with this and make it work it would be very informative!
Andy - 01A plate dissipation is not a problem. The plate is more than adequate size to dissipate 1.2 W at 8 mA. 26 has same plate size and construction as 01A, and it dissipates 1.1 W at its highest tabulated OP -in addition to 1.5 W filament. Actually, 01A at 135 V 3 mA is running too cool (barely warm to touch) for optimal gettering. Good gettering is extremely important for the longevity of its very thin TT filament, that's why it has, unlike most other tubes, getter flash all over the envelope. Thus, running at higher current, you kill two bunnies with one whammy - better OP and longer life.
01A data sheets don't have high current operating point(s) because it was deemed unnecessary for the ways the tube has been employed in 1920s. They cared most about reduction of current draw from the B+ battery.
01A data sheets don't have high current operating point(s) because it was deemed unnecessary for the ways the tube has been employed in 1920s. They cared most about reduction of current draw from the B+ battery.
The rule of thumb for a good low distortion driver is maximum rms output voltage in the ballpark of 10% of driver's DC plate voltage.
01A has mu=8. Gain equal to mu can be realized with transformer coupling.
I have plenty of 01A and this encourages me to try it at 6mA for instance. It certainly is a good sounding tube.
...almost...
Even -5mA- CCS loaded #01a capable 40.5Vpp at 2VRMS input.
Is this enough to drive any DHT power tube?
BTW #01a -in my practice, without any help: SF, CF- unable to normally drive any stronger capacitive load at 20kHz.
Last edited:
Very interesting thanks sser2!
I had similar thoughts, but Im far behind you. I haven't got it on the bench yet, still collecting a few components.
- Modern Radio Reception - has some good info and discusses 6mA with no bias
http://ia902704.us.archive.org/27/items/modernradiorecep00leutrich/modernradiorecep00leutrich.pdf
- Noted Ale discussed at 3-5ma
https://www.bartola.co.uk/valves/2020/12/29/01a-low-gain-dht-preamp/
- I am considering the VT62 which offers the gain Id like and appears relatively easy to drive. I assume @euro21 exhausted the 01A-801A idea, but I wanted to give it a shot!
- I intended to use balanced DAC with higher output (Im currently using a Topping D90SE capable of 5V , and use an input transformer to convert to single ended before input into Ale's Gyrator, https://www.audiosciencereview.com/forum/index.php?threads/topping-d90se-review-balanced-dac.24235/
MANY thanks for the additional details on the 01A! I cant wait to read more!
I had similar thoughts, but Im far behind you. I haven't got it on the bench yet, still collecting a few components.
- Modern Radio Reception - has some good info and discusses 6mA with no bias
http://ia902704.us.archive.org/27/items/modernradiorecep00leutrich/modernradiorecep00leutrich.pdf
- Noted Ale discussed at 3-5ma
https://www.bartola.co.uk/valves/2020/12/29/01a-low-gain-dht-preamp/
- I am considering the VT62 which offers the gain Id like and appears relatively easy to drive. I assume @euro21 exhausted the 01A-801A idea, but I wanted to give it a shot!
- I intended to use balanced DAC with higher output (Im currently using a Topping D90SE capable of 5V , and use an input transformer to convert to single ended before input into Ale's Gyrator, https://www.audiosciencereview.com/forum/index.php?threads/topping-d90se-review-balanced-dac.24235/
MANY thanks for the additional details on the 01A! I cant wait to read more!
VT62/801a requires about 80-100Vpp in A1 (depending of operating point), so any tube with -about- gain=8 (less than 45Vpp) unable to drive it without another gain stage (for example 1:4 SUT).
I drive it with 841+CF or SF (without it 841 also unable to drive properly), or 1:4 SUT + 801a gain stage.
#01a isn't enough robust to drive (without another driver stage) larger power tubes.
Euro21: what you say about 01A's ability to drive capacitive load at 20 kHz is like a bucket of cold water over my head. Especially from someone who extensively experimented with this tube.
But I am used to taking cold showers. I have an ace in my sleeve: a power DHT with amplification factor of 4 and grid to plate capacitance of only 5 pF. I will disclose it in due course.
One might also ask whether full power at 20 kHz is required in an amplifier that is intended for playing music. First, we never use amplifiers at full power; more typically it is just a small fraction of it. Second, music's energy is concentrated at lows and low mids, and progressively decreases towards highs. Third, many of us, myself included, do not hear beyond 10-12 kHz, which in no way interferes with our enjoyment of quality sound reproduction. With these points in mind, a more realistic, but fully adequate goal would be something like 10% of undistorted full power at 20 KHz.
But I am used to taking cold showers. I have an ace in my sleeve: a power DHT with amplification factor of 4 and grid to plate capacitance of only 5 pF. I will disclose it in due course.
One might also ask whether full power at 20 kHz is required in an amplifier that is intended for playing music. First, we never use amplifiers at full power; more typically it is just a small fraction of it. Second, music's energy is concentrated at lows and low mids, and progressively decreases towards highs. Third, many of us, myself included, do not hear beyond 10-12 kHz, which in no way interferes with our enjoyment of quality sound reproduction. With these points in mind, a more realistic, but fully adequate goal would be something like 10% of undistorted full power at 20 KHz.
Few years ago in one conversation I mentioned to Ale Moglia, that I have experienced that #01a unable to load even few hundred pF load (long coaxial cable) at HF, so HF behaviour was strange.
He discussed about it (slew rate problem) there:
https://www.bartola.co.uk/valves/2015/12/21/slew-rate-in-preamps/
Aside from that the #01a is great tube, but in her own place. If you acknowledge its limitations during designing and complete with appropriate driver (for example CF/SF), it's good low gain preamp tube.
As (most of) another ancient DHT tubes, it's gain not too much, so use it as (gain) driver is limited.
Very few DH tube has enough gain to use in power tube driving stage, and even less can do it as in two stage amplifier.
Even 20A enough strong (has enough low output impedance) to drive most of tubes ... but it's gain give limitation.
For example I'm "in love" with 841.
It's gain -mostly- enough to drive power tubes, but also has -strong- limitation (enormous load requirement), so any beefy driver (CF/SF) also need for proper using.
So no free lunch. :-(
He discussed about it (slew rate problem) there:
https://www.bartola.co.uk/valves/2015/12/21/slew-rate-in-preamps/
Aside from that the #01a is great tube, but in her own place. If you acknowledge its limitations during designing and complete with appropriate driver (for example CF/SF), it's good low gain preamp tube.
As (most of) another ancient DHT tubes, it's gain not too much, so use it as (gain) driver is limited.
Very few DH tube has enough gain to use in power tube driving stage, and even less can do it as in two stage amplifier.
Even 20A enough strong (has enough low output impedance) to drive most of tubes ... but it's gain give limitation.
For example I'm "in love" with 841.
It's gain -mostly- enough to drive power tubes, but also has -strong- limitation (enormous load requirement), so any beefy driver (CF/SF) also need for proper using.
So no free lunch. :-(
Few hundred pF... O horror.
I never understood the whole Idea of preamp connected with a cable to power amp. Few hundred pF cable loading high impedance source is an unnecessary self-inflicted wound requiring CF/SF remedies. Cable should be where it doesn't matter - interconnecting a very low output impedance (few Ohms) source, like CD player, with high input impedance integrated amplifier.
In an integrated amplifier, a relatively high (10 K) Rp tube CAN drive tens of pF at the input of the following stage. As calculated, 8 mA 01A IT-coupled, as suggested, to a 5 pF grid-to-plate, mu=4 triode, is equivalent to 13 mA triode, like 6SN7, directly driving 2A3. The latter has been proven to work. Whether my calculations are correct, we'll see.
I follow Ale Moglia's work and keep reading and re-reading his posts. I respect and admire him, although may disagree with some of his approaches.
I never understood the whole Idea of preamp connected with a cable to power amp. Few hundred pF cable loading high impedance source is an unnecessary self-inflicted wound requiring CF/SF remedies. Cable should be where it doesn't matter - interconnecting a very low output impedance (few Ohms) source, like CD player, with high input impedance integrated amplifier.
In an integrated amplifier, a relatively high (10 K) Rp tube CAN drive tens of pF at the input of the following stage. As calculated, 8 mA 01A IT-coupled, as suggested, to a 5 pF grid-to-plate, mu=4 triode, is equivalent to 13 mA triode, like 6SN7, directly driving 2A3. The latter has been proven to work. Whether my calculations are correct, we'll see.
I follow Ale Moglia's work and keep reading and re-reading his posts. I respect and admire him, although may disagree with some of his approaches.