Design Methodology: valve anode voltages - UK Vintage Radio Repair and Restoration Discussion Forum

Lee de Forest · 25th Feb 2009, 2:01 pm

I have been looking at some old circuits for AM radios and there are several which use the following Mullard valve line up with a 230V ac power supply;
ECH35, EF39, EBC33.
Although the designs use the same valves there is a large variation in the specified working anode voltages e.g. 80 volts to 250 volts for the ECH35.
Why such large variations in the design voltages for the same valve running from the same power supply

One idea which crossed my mind was that the lower voltage sets were low power output long life designs.

Oldtimer74 · 25th Feb 2009, 8:27 pm

Since I guess that the main HT supply for all these radios was about 250V how was it reduced for the circuit you said had 80V on the anode? Was this by purely a decoupling resistor in the HT line to the IF transformer, or was it stabilised for some reason? Many cheap set circuits would dispense with any component they could get away with, so the full HT would be applied to the mixer and IF valve anodes to save the cost of components. Business was VERY competitive in those days. If they could get away with it without getting feedback, for instance, they would. Of course they would have to put a resistor in the screen supply as this must be lower than the anode voltage. The better sets might feed the screens from an individual potential divider for each valve. The cheaper ones might use a single common resistor to drop the voltage for more than one valve.

Pat G3IKR

Lee de Forest · 26th Feb 2009, 2:12 pm

From memory the voltages were dropped using decoupling resistors in the HT line some of the circuits used the one resistor to feed the screens, some had individual resistors. I am interested in this because I am trying to understand why there was such a large variation in the working voltages for different designs. I had not considered the cost part in the design process but, as you say, they would reduce costs by eliminating any component they could get away with. It may only have been one resistor but if you could do that in several places with other parts it would soon stack up when you consider the reduction in fitting costs. I would liked to have been there to see how the design was developed to know the real reasons for doing things, I would imagine a lot of it was variation on a theme combined with some science and some rule of thumb.

Tim · 26th Feb 2009, 2:22 pm

Quote:

The cheaper ones might use a single common resistor to drop the voltage for more than one valve.

But would't they then need a larger(and more expensive) de-coupling capacitor?

Sideband · 26th Feb 2009, 2:39 pm

The other possibility is that there were variations with what the voltages were measured with. For example an early radio may have had voltages taken with an Avo 7. This would give lower readings than if the same voltages were measured with an Avo 8. Most circuits will tell you what the voltages were measured with. A typical example today would be for example measuring the anode voltage of a DAF96 battery pentode with a digital meter. You'd probably find the voltage was around 40v. Using an Avo 7 however on say the 500v range (that gives the highest impedance) it may only measure 20v.

Just a thought.

Rich.

Lee de Forest · 27th Feb 2009, 2:09 pm

That's a big difference, it would not have occured to me that there would be such a large deviation in measurements.
I will check the quoted resistances for the different circuits to see if the values would be theoretically lower.
I had a bit of a play last night with my home made set that uses this line up. The anode voltages were recorded to start with and then I fitted some resistors to reduce the voltages. As expected a reduction in voltage gave a reduction in volume but I found that up to a point a large reduction in voltage did not give that much of a drop in volume. e.g. 250v to 150v on the anode of the EF39. Reducing the voltage on the ECH caused some instability(resonance) but this was cured by fitting a capacitor between the anode of this valve and earth/chassis.
The size and ratings of other parts might effect the voltages used i.e. the size of the I.F. transformers!

LCTorana · 19th Mar 2009, 7:07 am

One reason that the converter and IF valves were run at such low voltages is that they can be run without any grid bias at all, reducing componentry and cost.

This can have the welcome side-effect of improving long-term reliability, both because there are less components to cause trouble, and because the effects of valve charecteristic drift are compensated for.

Lee de Forest · 19th Mar 2009, 1:52 pm

Thanks I had not thought about that. I have read recently about certain later versions of valves not needing cathode resistors and capacitors.
I will look at the old circuits again and compare different anode voltages for the same valve against grid bias circuitry.

25th Feb 2009, 2:01 pm	#1
Lee de Forest Retired Dormant Member Join Date: Sep 2008 Location: Harrogate, North Yorkshire, UK. Posts: 352	Design Methodology: valve anode voltages I have been looking at some old circuits for AM radios and there are several which use the following Mullard valve line up with a 230V ac power supply; ECH35, EF39, EBC33. Although the designs use the same valves there is a large variation in the specified working anode voltages e.g. 80 volts to 250 volts for the ECH35. Why such large variations in the design voltages for the same valve running from the same power supply One idea which crossed my mind was that the lower voltage sets were low power output long life designs.

25th Feb 2009, 8:27 pm	#2
Oldtimer74 Rest in Peace Join Date: Apr 2008 Location: Worcester, Worcestershire, UK. Posts: 330	Re: Design Methodology: valve anode voltages Since I guess that the main HT supply for all these radios was about 250V how was it reduced for the circuit you said had 80V on the anode? Was this by purely a decoupling resistor in the HT line to the IF transformer, or was it stabilised for some reason? Many cheap set circuits would dispense with any component they could get away with, so the full HT would be applied to the mixer and IF valve anodes to save the cost of components. Business was VERY competitive in those days. If they could get away with it without getting feedback, for instance, they would. Of course they would have to put a resistor in the screen supply as this must be lower than the anode voltage. The better sets might feed the screens from an individual potential divider for each valve. The cheaper ones might use a single common resistor to drop the voltage for more than one valve. Pat G3IKR

26th Feb 2009, 2:12 pm	#3
Lee de Forest Retired Dormant Member Join Date: Sep 2008 Location: Harrogate, North Yorkshire, UK. Posts: 352	Re: Design Methodology: valve anode voltages From memory the voltages were dropped using decoupling resistors in the HT line some of the circuits used the one resistor to feed the screens, some had individual resistors. I am interested in this because I am trying to understand why there was such a large variation in the working voltages for different designs. I had not considered the cost part in the design process but, as you say, they would reduce costs by eliminating any component they could get away with. It may only have been one resistor but if you could do that in several places with other parts it would soon stack up when you consider the reduction in fitting costs. I would liked to have been there to see how the design was developed to know the real reasons for doing things, I would imagine a lot of it was variation on a theme combined with some science and some rule of thumb.

26th Feb 2009, 2:39 pm	#5
Sideband Dekatron Join Date: Oct 2003 Location: Croydon, Surrey, UK. Posts: 7,572	Re: Design Methodology: valve anode voltages The other possibility is that there were variations with what the voltages were measured with. For example an early radio may have had voltages taken with an Avo 7. This would give lower readings than if the same voltages were measured with an Avo 8. Most circuits will tell you what the voltages were measured with. A typical example today would be for example measuring the anode voltage of a DAF96 battery pentode with a digital meter. You'd probably find the voltage was around 40v. Using an Avo 7 however on say the 500v range (that gives the highest impedance) it may only measure 20v. Just a thought. Rich. __________________ There are lots of brilliant keyboard players and then there is Rick Wakeman.....

27th Feb 2009, 2:09 pm	#6
Lee de Forest Retired Dormant Member Join Date: Sep 2008 Location: Harrogate, North Yorkshire, UK. Posts: 352	Re: Design Methodology: valve anode voltages That's a big difference, it would not have occured to me that there would be such a large deviation in measurements. I will check the quoted resistances for the different circuits to see if the values would be theoretically lower. I had a bit of a play last night with my home made set that uses this line up. The anode voltages were recorded to start with and then I fitted some resistors to reduce the voltages. As expected a reduction in voltage gave a reduction in volume but I found that up to a point a large reduction in voltage did not give that much of a drop in volume. e.g. 250v to 150v on the anode of the EF39. Reducing the voltage on the ECH caused some instability(resonance) but this was cured by fitting a capacitor between the anode of this valve and earth/chassis. The size and ratings of other parts might effect the voltages used i.e. the size of the I.F. transformers!

19th Mar 2009, 7:07 am	#7
LCTorana Triode Join Date: Mar 2009 Location: Melbourne, Australia Posts: 17	Re: Design Methodology: valve anode voltages One reason that the converter and IF valves were run at such low voltages is that they can be run without any grid bias at all, reducing componentry and cost. This can have the welcome side-effect of improving long-term reliability, both because there are less components to cause trouble, and because the effects of valve charecteristic drift are compensated for.

19th Mar 2009, 1:52 pm	#8
Lee de Forest Retired Dormant Member Join Date: Sep 2008 Location: Harrogate, North Yorkshire, UK. Posts: 352	Re: Design Methodology: valve anode voltages Thanks I had not thought about that. I have read recently about certain later versions of valves not needing cathode resistors and capacitors. I will look at the old circuits again and compare different anode voltages for the same valve against grid bias circuitry.