Low Voltage Tube Pre - How to calculate andode voltage?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

JanusRec

Well-known member
Joined
Oct 25, 2008
Messages
66
Location
Berlin, Germany
Hi everybody,

I have a really nic tube pre design from a german pro audio magazin using a EF98 pentode and optinally a ECC86 double triode for an alternative sound or as a second stage. My plan is to switch the tubes with relays (already designed a pcb and everything) but I left them on the schematic to keep it simple:

k2extrakt.gif


So here are my questions:

As you can see the original was designed for +24V supply voltage (they made the thing for a switched wall supply) and drop it to soemething like 6 to 10V with R22, R23 and R30. But I'm going to use a linear supply with about 15V or 18V, so I can use the same power rail for the tubes and the output stage which consists of a OPA2134 and a DRV134.
1. How do I calculate R22, R23 and R30 and what kind of voltage drop is happening in the tube, respectively? I know ohm's law, but as it seems that's not sufficient for this case  :D

In the audio mag it sais you can adjust the gain of the EF98 via R29.
2. How do I calculate the voltage gain? I guess it has something to do with biasing and the katode voltage, but in the data sheet I found only charts for the anode current... So I'm ignoarant in how to use tube data sheets. Please illuminate me ;-)

Thanks for your help,

Sebastian
 
i am of no help, but could you post a little more info on this project? such as the output stage circuit, and the recommended input transformer?
thanks!
;D
 
Hi Janus,

Welcome to the forum seeing as I haven't seen you write here before.

Anyway, I think you should read this:  http://www.audioxpress.com/resource/audioclass/ga699ac.pdf

There may be something equivalent in German somewhere.

Basically, you need to find out the idle current of the valves in order to determine how much voltage drop will occur over those plate resistors. The values of these resistors also determines gain and distortion.
 
dougdeeper said:
i am of no help, but could you post a little more info on this project? such as the output stage circuit, and the recommended input transformer?
thanks!
;D

Input transformer is a Pikatron 1:15 (Üp3096M) and the output stage consists of a half OPA2134 for impendace matching an a DRV134 line driver. I understad every part of the schematic execpt of the tube part, so no problem there. I can post the whole circiut later, if you're interested.

rodabod said:
Hi Janus,

Welcome to the forum seeing as I haven't seen you write here before.

Anyway, I think you should read this:  http://www.audioxpress.com/resource/audioclass/ga699ac.pdf

There may be something equivalent in German somewhere.

Basically, you need to find out the idle current of the valves in order to determine how much voltage drop will occur over those plate resistors. The values of these resistors also determines gain and distortion.

Thanks for the link, English is fine.
Yes, haven't posted to often, but if've been reading a lot and built a couple of projects. Just only posted for touble shooting once or twice.
 
JanusRec said:
Hi everybody,

I have a really nic tube pre design from a german pro audio magazin using a EF98 pentode and optinally a ECC86 double triode for an alternative sound or as a second stage. My plan is to switch the tubes with relays (already designed a pcb and everything) but I left them on the schematic to keep it simple:

k2extrakt.gif


So here are my questions:

As you can see the original was designed for +24V supply voltage (they made the thing for a switched wall supply) and drop it to soemething like 6 to 10V with R22, R23 and R30. But I'm going to use a linear supply with about 15V or 18V, so I can use the same power rail for the tubes and the output stage which consists of a OPA2134 and a DRV134.
1. How do I calculate R22, R23 and R30 and what kind of voltage drop is happening in the tube, respectively? I know ohm's law, but as it seems that's not sufficient for this case  :D

In the audio mag it sais you can adjust the gain of the EF98 via R29.
2. How do I calculate the voltage gain? I guess it has something to do with biasing and the katode voltage, but in the data sheet I found only charts for the anode current... So I'm ignoarant in how to use tube data sheets. Please illuminate me ;-)

Thanks for your help,

Sebastian

The only resistor that is dropping voltage to the plate of the pentode is R22. If the B+ is at 24V and the plate is at 10V, then the resistor is dropping 14V across it... 15k resistance, use ohm's law to determine current through the plate resistor.

The bias point for a pentode is much more complicated than a triode because the screen voltage (here the screen grid is connected to the junction of R23 and C20) actually determines the grid curves, instead of worrying only about plate voltage, in this case you also need to know screen voltage. The plate:screen current ratio is usually pretty close to fixed, that is, until your plate voltage swings down towards 0V, at which point the screen current generally increases.

I haven't looked at the tube data sheet here, but I'm guessing this stage is set up a little unusually in that the screen might actually be at a higher voltage than the plate. This is assuming that the plate to screen current ratio is somewhere in the 2.5:1 to 7:1 range which is common for most the pentodes I'm familiar with. The plate and screen resistors are pretty close in value.

It would be easier to analyze the circuit if you could provide voltages.

BTW this sounds like an interesting project.

mason
 
The pentode cathode-screen drop is essentially the Mug2 times the Cathode Resistor. Plate pulls some multiple of screen current. So for the small silly change you propose, everything changes changes in proportion. If the circuit works at 24V (however poorly), then it works at 15V (more poorly). No change needed.

> it sais you can adjust the gain of the EF98 via R29.

+/-3dB. Since I think the gain of this amp will always be way-wrong, it is hardly worth implementing.

> How do I calculate the voltage gain?

It really IS easier to build it and measure it.

If it were biased "well", gain is about 6. However it appears to be saturated, plate stuck low.

> already designed a pcb

Sorry to hear that. It surely is not happy the way it is drawn here. There is no gridleak resistor on the triode stage. Also you would expect some cathode bias.

I really think you ought to breadboard it.
 
rodabod said:
Anyway, I think you should read this:  http://www.audioxpress.com/resource/audioclass/ga699ac.pdf

Ok, I read it and I think I understood it more or less. But they don't really explain how they chose their plate-load resistor. They just pick a 100K resistor without explaining why and then start calculating.
Or is that the way you do it: Picking some supply voltage example from the tube manual and then randomly try plate-load resistors until your load-line looks nice? That was for triodes though. The pentode part in that paper is really short and simplyfied (I guess) and they use tables with plate-load resistors and voltages that I don't have in my data sheet. Any more links on that topic?

PRR said:
The pentode cathode-screen drop is essentially the Mug2 times the Cathode Resistor. Plate pulls some multiple of screen current. So for the small silly change you propose, everything changes changes in proportion. If the circuit works at 24V (however poorly), then it works at 15V (more poorly). No change needed.

Thanks for the grid resistor and biasing hint for the triode. I will look to that later. And probably I'm going to use the triode as the first stage - less distortion, right? That at least what I read in the tube paper from the beginning of the post.
What is Mug2? I found it in the data sheet, but I have no idea what it is and didn't find anything on the net.
And why do you the think the cicuit works poorly? It's a low-voltage car radio tube, so there should be a way to make it work properly at 12,6V or 25V.

Oh, and I didn't etch anything yet, so I can still make changes...
 

Latest posts

Back
Top