Hi. I've spent countless hours reading and experimenting with red plating, and my last option was to ask here, which seems like the most active forum in regards to tubes.
Firstly, this amp is different to what is generally discussed here. It's a 4-tube RF amp, not an audio amp, but I'm guessing or hoping that the solution may be something that applies to tube amps in general.
In case the answer isn't a simple one, I can go into more detail, else I'll try and keep it brief. At first there was always least one tube red plating, even with low input driving the tubes nowhere near what they're capable of. I've tried several tubes, all with similar results. There was originally one resistor to control the bias for all 4 tubes, but I now split the tracks for all 4 bias voltages, and can control each one separately. Currently it seems it's all quite equal, as
they're red plating equally lol. I've tried -8volts (seems fine for the output) and -50volts (which drops the output a bit too much) to no avail.
The tubes in question are 6P45S, which I've been told apparently require some sort of auto-bias system.
What I wish to know is, where to from here, any suggestions? Is there a general rule to prevent all tubes from red plating at the same time?
Thanks.
Firstly, this amp is different to what is generally discussed here. It's a 4-tube RF amp, not an audio amp, but I'm guessing or hoping that the solution may be something that applies to tube amps in general.
In case the answer isn't a simple one, I can go into more detail, else I'll try and keep it brief. At first there was always least one tube red plating, even with low input driving the tubes nowhere near what they're capable of. I've tried several tubes, all with similar results. There was originally one resistor to control the bias for all 4 tubes, but I now split the tracks for all 4 bias voltages, and can control each one separately. Currently it seems it's all quite equal, as
they're red plating equally lol. I've tried -8volts (seems fine for the output) and -50volts (which drops the output a bit too much) to no avail.
The tubes in question are 6P45S, which I've been told apparently require some sort of auto-bias system.
What I wish to know is, where to from here, any suggestions? Is there a general rule to prevent all tubes from red plating at the same time?
Thanks.
In class C and with no input the tubes should be cut off and run "cold".
Or did I miss something in your message ?
Sorry, not sure. I haven't measured any current, and Im unfamiliar with the terms Ug2 and Ua. I wasn't really keen on having to learn the ins and outs of tubes, but seeing as I've bumped into this issue, it's become a bit of a project to work out, but I haven't given up just yet.
To explain the bias voltage - originally the tubes were PL519's, which require 40 volts for the heater. This voltage is also used for the bias, via a diode and cap to convert it to DC.
When I converted to the 6P45S's, which require 6.3 volts for the heater, rather than try to drop the 40 volts AC down with a resistor (which I believe would've been too unstable), I added another transformer with 6.3 volts AC output. That output is also used for the bias. However, I also have the option of using the now unused 40 volt line, which dramatically reduces the output, but doesn't eliminate the red plating of the tubes.
Most class C RF tube based amplifiers, get the cutoff voltage from grid rectification from the input signal, drawing some amount of power from preceding stage, particularly if that previous stage is in the same equipment. In case of abnormally low signal input voltage, the tubes get only low bias and then gets red plates. The same can occur if the tuned circuits at the plate of the tube aren´t tuned to the incoming signal, or a integer multiple of such frequency, in that case the stage is a frequency doubler, tripler, etc., although signal level is right. But if it is independent set, may be biased externally by means of a -C power supply. I don´t know exactly the case of the guy is making the question.
Sometimes the term "linear" is misunderstood. A power amplifier for FM can be class C, and also called "linear" when in true it isn´t.
The PDF can be accessed by goint to www.rmitaly.com then >download >download manuals >KLV1000-manual_rel_210.pdf
According to manual the original tubes are PL509 having 40 V heater voltage, which is also used to generate the bias voltage.
Now it seems that tubes have been changed to 6P45S, which has 6,3 V heater voltage. So what has happened to bias voltage, which should be much more than just rectified 6,3 VAC can form.
The simple fact is that if tubes are redplating they are dissipating too much power/heat. This reguires high quiescent current which in turn can take place if the bias voltage is too low.
Now it seems that tubes have been changed to 6P45S, which has 6,3 V heater voltage. So what has happened to bias voltage, which should be much more than just rectified 6,3 VAC can form.
The simple fact is that if tubes are redplating they are dissipating too much power/heat. This reguires high quiescent current which in turn can take place if the bias voltage is too low.
Thanks for the responses. I understand about the bias voltage being too low, so I removed one end of the diode(s) and applied the 40 volts AC direct (leaving the 6.3 volts AC intact for the heater), which read about 50 volts DC on the other end of the diode, for the bias.
The same thing still occurs. I can turn the drive right down to the point where there is very little output, and they all still red plate. They take a little longer to do so, but it still happens.
Neutralizing.
Does the driver also gets hot plates?. May be that the problem is a VHF self oscillation, at a frequency where the plate circuit can´t absorb it. Make an easy RF power meter: a Hertz ring. Take a small incandescent lamp, the less power as possible will result in a higher sensitivity. Make a wire loop of about 50 mm (2"), (1 turn coil) and solder both ends to each pin of the lamp. Carefully, put the loop near the output LC tank, and see if with no input, the lamp brights. If this is the case, try to re-neutralize all set. Search in the web how to neutralize it, or post me and I try to explain it. It´s easy.
Also, tell me if the trouble started from a tube replacement. It points to a neutralizing trouble.
Does the driver also gets hot plates?. May be that the problem is a VHF self oscillation, at a frequency where the plate circuit can´t absorb it. Make an easy RF power meter: a Hertz ring. Take a small incandescent lamp, the less power as possible will result in a higher sensitivity. Make a wire loop of about 50 mm (2"), (1 turn coil) and solder both ends to each pin of the lamp. Carefully, put the loop near the output LC tank, and see if with no input, the lamp brights. If this is the case, try to re-neutralize all set. Search in the web how to neutralize it, or post me and I try to explain it. It´s easy.
Also, tell me if the trouble started from a tube replacement. It points to a neutralizing trouble.
Yes. I just noticed today that the driver also gets hot plates. Another thing I've noticed is the output is always slowly rising. Say for example with about 3 watts in I get about 300 out, after about 30 seconds or so that might rise to ~350, and keep rising.
I'm not sure if I've always had this trouble, as I was unaware of red plating until the first time I had seen it a while ago. I have several of these tubes and they all respond the same way.
I don't think I have a low voltage incandescent lamp on hand, but if neutralizing is an easy task, I can try that if you can explain.
Thanks.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.