High filament voltage

[Demsy]

I've been running the 2A3 Paramount on 255v line voltage as suggested. The 2A3 filament voltage is 2.4v which is ok. But the 5670's filament is then 6.7v, measured between pin 1 and 9. And to be frank, I don't like this high 5670 filament voltage.
It was mentioned that there is nothing wrong with the 240v transformer, but looking at the numbers, obviously something is not right. My 5670's are glowing rather bright, I'm afraid with this voltage the lifetime will suffer. I'm using the NOS Tungsol 5670, not really hard to get, but also not cheap.
I checked the voltage list, page 67, of the Paramount v 1.1 Manual, the voltage should be around 3.1v on pins B1 and B9. I know that there is a certain working voltage for the filament, if I'm not wrong it's +\- 10%. But it is always better to stay under the nominal of 6.3v, for the lifetime's sake. And I believe that is also the reason why the pin 1 and 9 voltage is kept at 3.1v.
When I ordered the 240v version, I had to pay $30.00 extra for the transformers. But with this experience I think I'm better off with the original 120v transformer and use a stepdown unit instead. This way I don't have to make all the changes that are recommended in PJ's earlier reply. My dilemma now is, what to do......get the 120v transformers?
Any comment, Doc, PJ or PB, especially about the 240v transformer?

[Paul Birkeland]

Having 6.7V instead of 6.3V is 6% over target, and certainly within design parameters specified in the datasheet (6.9V max).

-PB

[Doc B.]

Yup, I agree, those numbers sound fine.

[Paul Joppa]

The secondaries of the 120v transformer are identical to those of the 240v transformer. We have checked this in the lab with parts from our production stock. Changing transformers will not change your results.

As noted by others, the design is operating within spec. Nevertheless if you want to trim things more closely, the simplest solution would be to install a 1.1-ohm resistor in series with the heater supply to the 5670, or perhaps a 0.56 -ohm resistor at each pin.

[Demsy]

The secondaries of the 120v transformer are identical to those of the 240v transformer. We have checked this in the lab with parts from our production stock. Changing transformers will not change your results.

As noted by others, the design is operating within spec. Nevertheless if you want to trim things more closely, the simplest solution would be to install a 1.1-ohm resistor in series with the heater supply to the 5670, or perhaps a 0.56 -ohm resistor at each pin.

Paul, you mentioned that changing the transformer will not change the results. I don't understand this. My complain is that in order to reach the designed filament voltage of the power tube, 300B or 2A3, the line voltage should be higher than 240v. Ideal, in my case is up to 260v. Which in turn will increase the filament voltage of the driver tube above the recommended 3.1v. I take it that this is not necessary with the 120v transformer, so what do you mean that changing the transformer will not change the results?? You're not saying that this is also the case with the standard supplied 120v transformer, are you? All what's needed in the case of the power transformer is that the transformer will supply a given output voltage/s with a certain input voltage, simply put. Please explain this as I really do not understand it, to me something is not right with the 240v transformers that I have.

As for the high filament voltage of the driver tube, I sincerely believe that this has an impact on how the amp sounds and also on the lifetime of the tube itself. Unfortunately I cannot compare it, if the voltage of this tube is right, than the power tube's filament voltage is really low. Which results in a compressed lifeless sound. I have built the Hagerman Cornet2 phono preamp many years ago and I remember that the change/s of the filament voltage/s strongly influence the sound, I'm not surprised if this is also the case with the 5670. PB's stickied posting and my own experience confirmed this, although it's on the 300B/2A3.

To bring this filament voltage down, what wattage of the resistor is needed, Paul? I will immediately try to get the resistors and try it out.

[Doc B.]

Simply square the current draw of the 5670 heater and multiply that by the resistor value. (.35)^2 X1.1 = .135W. Triple that and you will have a power rating value with a good safety margin, .4W. A 3/4W or 1W rating will be great.

[Demsy]

Thanks, Doc.

[Paul Joppa]

Doc B got this (thanks!) - sorry I didn't include the power spec.

Yes, I am saying we have an identical situation with the 120v transformer. Probably it has not come up because most 120v power is 120 to 125 volts. In 220 or 230 or 240v countries, outside northern Europe, the power is often below 240v.

We are working to better accommodate these voltage variations in future products.

[Demsy]

Ah, this also happens on the standard 120v transformer. Well then, I'll get and install the resistors.

[Demsy]

Installed the resistors , mains voltage at 260v. 5670 Tung-Sol filament 6.3v, JJ 2A3-40 filament 2.46v.

[mcandmar]

260v is high enough that your supplier would probably want to correct it, have you tried contacting them out of interest?

[Demsy]

260v is high enough that your supplier would probably want to correct it, have you tried contacting them out of interest?

Hi mcandmar,
If it is me that you mean, I did contact the Bottlehead about the transformer, didn't I.
Intentionally is the line voltage brought up to 260v in order to get the output tube filament voltage closer to 2.5v.

[Paul Joppa]

Misunderstanding - mcandmar was suggesting the power company might want to reduce a high 260v power line voltage. But Demsy has a low power line voltage, around 230v if I recall correctly, and has raised it with a variac to bring up the filament voltage.