Morgan Jones discusses elevating heater voltages with a voltage divider between B+ and 0v. He mentions RCAs recommendations that lead to a 40v difference without any reference that I could find.
My RCA manual (RC-26) mentions a range of 15 to 80 volts. I understand that we want to keep the heater positive with respect to the cathode, and I understand that the value selected must be less than the "Maximun Heater-Cathode Voltage" but that is a pretty wide range in many applications.
So, how does one calculate the proper value?
My RCA manual (RC-26) mentions a range of 15 to 80 volts. I understand that we want to keep the heater positive with respect to the cathode, and I understand that the value selected must be less than the "Maximun Heater-Cathode Voltage" but that is a pretty wide range in many applications.
So, how does one calculate the proper value?
I think the range given is to cover a wide range of cathode voltages, whether as a classic anode follower or phase splitter, cathode follower, or whatever. Therefore, the "correct" voltage is simply one that is sufficient enough to keep the heater/cathode elements reverse biased under all operating conditions, without exceeding the heater/cathode voltage rating. That gives a wide latitude, but then the heater/cathode voltage rating is fairly large to begin with, versus the typical voltage that the cathode of an AF amplifier operates at. In minimizing noise, the use of such bias has the greatest impact on low level preamplifier type circuits, versus say high signal circuits like an output stage driver.
Dave
Dave
The aim is for the negative peak of the heater voltage (assuming a positive bias of the heater CT, or humdinger) to be about 5-10V positive of the cathode voltage (which may be at 0V, or biased positive by some amount (and possibly with a higher peak).
The only experimental understanding of this I have come across is from Klemperer in 1936 (http://dalmura.com.au/projects/Heater cathode insulation performance.pdf), who observed that the dynamic resistance between heater and cathode increased significantly when voltage differential exceeded 5-10V. The VI characteristic around zero voltage difference is not of a 'diode' nature, even though that is what is often used in descriptions (it's more like an anti-diode).
Ciao, Tim
The only experimental understanding of this I have come across is from Klemperer in 1936 (http://dalmura.com.au/projects/Heater cathode insulation performance.pdf), who observed that the dynamic resistance between heater and cathode increased significantly when voltage differential exceeded 5-10V. The VI characteristic around zero voltage difference is not of a 'diode' nature, even though that is what is often used in descriptions (it's more like an anti-diode).
Ciao, Tim
I think this issue was mentioned in "Getting the most out of vacuum tubes". It seems that, on average, 40V gives greatest valve life (with respect to heater-cathode breakdown). The curve has a broad minimum around this value, so no need to worry about the exact voltage. The normal practice of biasing heaters to ground, so a little lower than the cathode voltage, is not too bad and is simple which is why it is normally used.
Thanks to all for the reading and information. My object is to get rid of the last bit of hum in this single ended amp. I can scope a small 60hz wave form .15mv p-p on the dual triode cathodes which I thought was from the ripple in the DC LT supply leaking to the cathode. I elevated the heater voltage negative to 20v today (about 10 above the cathode) and saw no measurable benefit. I might be barking up the wrong tree.
Tim,
I only tried it with the probe connected to the cathode at the socket. That was a mistake. That waveform is present at the 0V side, at the ground plane and even at the power socket ground.
I was operating the amp off an isolation transformer and a Variac. When I plugged the amp to the wall directly the waveform dropped by a factor greater than 10. The hum at the speaker did not change. The waveform could be just as easily coming from the HT power supply as the LT.
So, I AM barking up the wrong tree with the hk leakage theory.
But, it's good to understand the elevated heater aspects more completely.
I only tried it with the probe connected to the cathode at the socket. That was a mistake. That waveform is present at the 0V side, at the ground plane and even at the power socket ground.
I was operating the amp off an isolation transformer and a Variac. When I plugged the amp to the wall directly the waveform dropped by a factor greater than 10. The hum at the speaker did not change. The waveform could be just as easily coming from the HT power supply as the LT.
So, I AM barking up the wrong tree with the hk leakage theory.
But, it's good to understand the elevated heater aspects more completely.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.