Heater voltages

[Telewire]

I'm building a bench HT PSU which needs to have a separate heater output for use in experimental projects.
This a variac-transformer arrangement which will supply up to 450v, the variac being connected to the primary of the HT transformer. This means that the heater winding of that transformer cannot be used because it's voltage will be varying in proportion to the variac supply. I therefore need another separate heater supply.

The problem is that dedicated 6.3v filament transformers seem to be expensive for what they are.
How critical is the heater voltage? A 6v supply would be far cheaper to arrange. Will the slight reduction make any noticeable difference to anything?

Thanks in advance

- Martin

[Skywave]

Most 6.3 v heater valves will operate quite O.K. with +/- 5% variation on this nominal 6.3 v. (The exception being really large power valves). Therefore, 6.0 v is just on the edge of the lower limit. However, I see that you state that this PSU is for "experimental use". Therefore, with a 6.0 v supply at source, there is no spare voltage for losses in the wiring to the Item Under Test - which, if it draws a substantial heater current, could pull down the supply to an unacceptably low figure (measured at the valve base pins), depending on the current capability of the 6v transformer.

One easy solution would be to fit an ex-valve radio HT transformer and just use its 6.3 v winding. If there are several 6.3v windings, you could bring them all out seperately. This also provides an arrangement for 12.6 v, which could arise as a need.

Moreover, since you require a PSU for experimental use, I would not favour the Variac-to-mains-transformer-primary design choice. As the voltage to the transformer primary is decreased by the Variac, the input VA to the transformer goes down accordingly. Thus the maximum current available from the transformer secondary goes down too. As the d.c. O/P voltage is decreased by the Variac, the regulation of the d.c. O/P will decrease. For experimental use, this characteristic is undesirable.

For an experimental PSU, far better design approach is to choose a linear series regulator valve with its accompanying error amplifier and voltage stabilizer valve as the reference. Such designs are common: the 'Net and Amateur Radio publications are good sources.

Hope that helps.

Al / Skywave.

[Boom]

In my experience the .3 of a volt can make a lot of difference if a device is only just making it due to low emisson. For test gear especially I would expect 6.3V and nothing less.

If a 6V transformer is easier to obtain then a tapped primary dropped down a step to 220-230 might bring the secondary up to 6.3V

[Telewire]

Thanks Chaps.
As the emphasis here is on 'cheap' , then another HT transformer, using the heater windings is the order of the day.
I can see what you mean about 6v being on the edge of the limit for this, especially if one has a tired valve in circuit (one gets a gut feeling-or doubt- about these things- which is why I asked).

Point taken about the current roll-off with the variac. Will do some tests. Suspect that a multi-tapped secondary transformer without the variac would be the obvious answer instead. This would also solve the original problem about heaters if there is a heater winding available....

-Martin

[kalee20]

Originally Posted by Skywave

As the voltage to the transformer primary is decreased by the Variac, the input VA to the transformer goes down accordingly. Thus the maximum current available from the transformer secondary goes down too.

Well, the VA available will decrease, but the output current available will stay the same! If you have a 25VA transformer, 250V @ 100mA, and you operate at half input voltage - you'd be able to run it at 12.5VA, and this is 125V 100mA. The current's stayed the same!

This is exactly what you'd expect, as the current rating of a transformer is determined by how much current the wire can carry without overheating.

In fact, in practice, at lower voltages you may be able to pull MORE current from the transformer - because at low voltages, the magnetic flux in the core will be less, so the core losses will be less, so the transformer won't get as hot. So, you have a bit more available loss due to winding resistance losses, to give the same overall temperature rise. Personally, I wouln't rely on up-rating current like this with an unknown transformer, but it does show you're going to be on the right side of safe.

As for heater voltages, for an experimental set-up I'd go for a 6.6V supply. Anything experimental is going to have longer leads, croc clips, terminals, etc and these can easily lose a few tenths of a volt. Even if not, you'll have shorter warm-up times, great for frequent on/offs on the bench.

Many small "6V" transformers aren't going to have brilliant regulation in any case, and you may find that operated below its maximum rating the voltage is higher. I've often found that "6.3V" heater transformers are larger, probably to give the better regulation and dependable output voltage that their intended application requires.

Originally Posted by Skywave

For an experimental PSU, far better design approach is to choose a linear series regulator valve with its accompanying error amplifier and voltage stabilizer valve as the reference.

This'd be the Rolls-Royce ideal to be aimed at, with maybe a panel knob to give a 10% variation, to allow for voltage drops as mentioned above.

[AlanBeckett]

Martin,
I would also suggest that a problem with 'cheap' 6v transformers is that the insulation will only be good for mains voltage, whereas you may be running the secondary of your transformer (eg regulator valves) at a much higher voltage. I've bought a couple of old RS heater transformers, meant as 'boost' transformers for CRTs, at very reasonable prices.
Alan

[Skywave]

Originally Posted by kalee20

Well, the VA available will decrease, but the output current available will stay the same! If you have a 25VA transformer, 250V @ 100mA, and you operate at half input voltage - you'd be able to run it at 12.5VA, and this is 125V 100mA. The current's stayed the same!

This is exactly what you'd expect, as the current rating of a transformer is determined by how much current the wire can carry without overheating.

Clearly, a reply from me is due on this. I have been away from home for a few days, else I would have responded earlier.

Simply put, Kalee is correct (of course); I was mistaken: a mental slip. I apologise to the OP and to any readers who were misled by that part of my Post.

Al / Skywave.