Yet another Transformer question!

[MrElectronicman]

I am currently playing about with a homebrew valve audio amplifier, but I am stuck because I need higher HT voltage. The transformer I an using only provides about 250 VDC and I would like something like 300+.

I have several transformers in my junk box but they are all low voltage types. However, I found an old transformer from the UK that is 240 VAC input and 17 VAC output. I also have another transformer that is 120 VAC to 16 VAC, so connecting them back to back I can get about 380 VDC no load when rectified, from a 120 VAC supply.

So far so good. The 120 VAC transformer is rated at 35.5 VA and is capable of supplying 2.2A according to the label. The 240VAC transformer has no indication of it's power rating, but looking at the flying leads from the secondary they are very thin, so I expect that is not rated very high.

Thus the question: Is there a way to check what current I might expect to get out of the 240 volt side of the transformer arrangement?

A quick guesstimate might say that the 17 volt secondary could supply 0.5 Amp, thus the output might be 9 VA, expecting the transformer to be about 50% efficient the input would be 20 VA, then the input current would be about 80-90 mA, at 240 Volts

Running the whole thing in reverse would be less efficient so I might expect 60-70 mA out. I am trying to see how much I could expect as I do not want to burn out the 240 Volt transformer.

Is my logic correct? or am I nuts

I will be getting a bigger power transformer, so this will not be the final arrangement, it's just for an experiment.

[G8HQP Dave]

Transformers are typically more like 90% efficient, generally better for bigger ones.

[peteronnith]

Hi David, estimating the current rating of transformers is a bit like guesswork, but try it and keep an eye (hand !) on the trannys to make sure they don't overheat. If your only feeding a push pull amp you should only need about 80/100 Ma. Should be OK.

Peter

[kalee20]

If the transformer is 9VA, then trivially working 'backwards' you'd be able to draw 38mA from the high voltage winding.

As G8HQP says, transformers are fairly efficient - thus the windings will be sized for the current they'll be expected to carry. It doesn't matter whether they are connected as intended, or in reverse.

Generally, efficiency is better in the intended direction than in reverse. But that's a small matter. The reason is that the input winding will carry magnetising current plus reflected load current, and the output winding will carry load current only. So the input (primary) winding will generally be wound with a wire gauge that makes it occupy slightly more than half of the space available, and the secondary slighly less than half. This gives the lowest total losses.

[Ed_Dinning]

Hi David, as a general rule a transformer is not damaged by small overloads and simply gets hotter. The heat is not immeadiatly noticed as it takes time to "come to the surface" so to speak.
As a rule of thumb it should not get to hot to easily keep your hand on; any hot smells or dripping wax and turn off at once.

You idea should work, you could also consider "adding" an additional winding in series with the HT winding, of the necessary voltage if you are using a bridge rectifier: or in series with the centre tap if it is a FW valve type.
At these voltages is should work fine as you will not be exceeding operating voltages and hence insulation levels by much.

It is not usually that easy to assess the current rating of a transformer winding. It can be done if you can measure the winding wire and use 3A per mm^2 as the current density, but this is a ball park figure only.

Ed

[TrevorG3VLF]

A transformer is normally wound to give a somewhat high voltage so that when loaded, the voltage is correct. If the transformer is run in reverse, the voltage will be low and will drop further under load. With transformers in series, you of course have two load drops

You may be able to get some idea of power rating from the size.

[MrElectronicman]

Thanks for the replies, I feel better about the arrangement now.

I have a couple of identical transformers connected back to back to make a double isolated arrangement and I put a small lamp connected to the low voltage side so I can see how much the loading is by the brightness of the lamp. I am not worried about damaging the transformers in that case as they are fairly robust.

Ed mentioned "adding" a winding in series with the HT winding. If I understood right, I had tried that technique many years ago with a couple of transformers trying to get an EHT voltage for a Oscilloscope. It was a bit of a disaster as the transformers began to spark as the insulation broke down, but then I was trying to get 2000 volts in that case.

I will try loading the 380VDC output carefully to see how much I can pull. If the voltage drops precipitously then I will have an idea of the current I can reasonable draw without possible damage.

[MrElectronicman]

Well, all my attempt to increase the HT voltage have come to naught. The double transformer trick could not produce more than about 20-25mA.

In thinking further about Ed Dinning's suggestion, I realised that he was talking about inserting additional voltage into to HT transformer center tap. I used another transformer that gave me about 56 VAC and put that into the center tap, but I could only get about 10 volts more on the DC.
The no load HT from the power supply was about 340 VDC, inserting the extra 56 VAC resulted in a no load voltage of 400 VDC. But when I connect the amp the HT voltage was only about 230 VDC under load, which was only 10 volts additional.

I guess I will have to get that bigger transformer a bit sooner.

[Ed_Dinning]

Hi David, yes the basic theory works but you also need to take the transformer winding impedances (NOT resistances alone) into account.

Ed

[glowinganode]

Hi David,
am correct in thinking the output drops from 400 Vdc no load, to 230 Vdc with only 25 mA of load?
If so it sounds like you need to check the smoothing arrangements, ie too small or failed filter capacitor.
As an aside, connecting the secondary of an LV transformer between the c.t. of the HT winding and 0 V, has the effect of moving the centre tap point.
So it now becomes a 370-0-330 transformer (for example). This may not matter with a valve rectifier, however may cause problems with solid-state.
Indeed the peak, no load output voltage will be increased however the mean on load output will be only slightly increased.
Rob.

[dinkydi]

If you are connecting a voltage to the centre-tap point to increase the power-supply output, that voltage should be rectified. If an AC voltage is connected, as you have tried, then FW rectification effectively becomes HW rectification because of voltage imbalance, as Rob pointed out. Regulation is then poor, as you subsequently discovered.

To get increased output from your power supply, the transformer voltage applied to the centre tap should first be rectified with a diode bridge, for example. Smoothing capacitors are not needed at the centre tap because of smoothing at the power-supply output.

Peter

[MrElectronicman]

Originally Posted by dinkydi

To get increased output from your power supply, the transformer voltage applied to the centre tap should first be rectified with a diode bridge,
Peter

Yes, you are absolutely right, I did not think of that. This way the transformer impedance has less effect since the voltage is mostly DC.

Originally Posted by glowinganode

am correct in thinking the output drops from 400 Vdc no load, to 230 Vdc with only 25 mA of load?
If so it sounds like you need to check the smoothing arrangements, ie too small or failed filter capacitor.
Rob.

The load current is about 90mA, the smoothing capacitors and the choke are all new so I do not expect them to be a factor. The real problem is that the HT transformer is only rated at 75mA.

Time to get the soldering iron warmed up again

[MrElectronicman]

Starting to make some headway. With a bridge rectifier installed I get 265 VDC with a load current of 100mA.
I have another transformer that produces about 40 VAC, so I will include this into the set up and I hope to get close to 300 VDC. I am a bit concerned that the load current will then be around 115mA, but for a short test I am expecting that this will be OK.

I will turn on the main transformer to let the valves heat up and then turn on the secondary transformers afterward, then the no load HT will not go above the rating of the smoothing capacitors (500V).

If you see a bright flash in the night sky, then you will know that I have failed

[MrElectronicman]

My experiment was not a total failure, managed to get 295 volts DC at 114mA but something was smelling rather hot so I quickly shut it all down. I was able to see that the higher voltage reduced the clipping at high volume levels however.

Now for another question:

In looking at the power transformers available from Hammond, they all have two low voltage heater windings, one 6.3VAC at various Amps, and another at 5VAC at 1-2 Amps. I am planning to use an EZ81 rectifier in the next PSU and would prefer to to able to use a separate 6.3Volt supply for the heater, obviously this will not be possible with the transformers available.

Are there any constraints on using a 6.3VAC heater supply to feed all the valves including the EZ81, or should I just get a small 6.3VAC transformer for the EZ81 only.

I have some other octal based rectifiers that use 5VAC for the heaters, but I would really like to use the EZ81.

[ColinTheAmpMan1]

Originally Posted by MrElectronicman

In looking at the power transformers available from Hammond, they all have two low voltage heater windings, one 6.3VAC at various Amps, and another at 5VAC at 1-2 Amps. I am planning to use an EZ81 rectifier in the next PSU and would prefer to to able to use a separate 6.3Volt supply for the heater, obviously this will not be possible with the transformers available.

Are there any constraints on using a 6.3VAC heater supply to feed all the valves including the EZ81, or should I just get a small 6.3VAC transformer for the EZ81 only.

I have some other octal based rectifiers that use 5VAC for the heaters, but I would really like to use the EZ81.

As far as I was aware, the EZ81 has particularly good heater-to-cathode insulation for just that reason. It can, and often is, run off the same 6.3VAC heater supply as the other valves. Mind you, this was probably done for cheapness in the transformer requirements. You ought to check that the 1.0 Amp requirement of the EZ81's heater isn't going to stress the mains transformer's heater winding and that the maximum rectified current of 150mA at 350-0-350 Volts is sufficient for your needs. The EZ80 only needs 0.6 Amp for the heater, but its max rectified current is only 90mA.

Regards, Colin.

[Mr Moose]

Hello,

To quote from the Brimar data sheet for the EZ81 ( http://frank.pocnet.net/sheets/155/e/EZ81.pdf )

"The BRIMAR EZ81 is a minature noval-based full wave rectifier for use in radio receivers and amplifiers using a common heater supply."

The peak Heater-Cathode potential is given as 500 volts.

Yours, Richard

[kalee20]

Yes, the EZ81 is designed to do just that.

Having said that, it is still good practice to minimise heater-cathode insulation stress when possible - a separate winding for the HT heater is better than a common heater. But if you haven't got one, then you haven't got one.

[Ed_Dinning]

Hi David, if there is space between the winding and the lams you can always add your own rectifier heater winding. (Best to use PVC insulated wire though).

Ed

[glowinganode]

Or add a few turns, as Ed suggests, to bring the 5 V winding up to 6.3 V.
PTFE (Teflon) insulated wire is particularly good for this purpose as the insulation is generally thinner, withstands higher temperature and is easier to thred through being slippery.

Rob.

[MrElectronicman]

Thanks for all the replies.

I had not noticed that the Heater to Cathode voltage was specified at 500Volts, make me feel a lot better now. I was more concerned about the max capacitor size that could be used.

Not sure about being able to thread a few turns of wire around the windings though. I have not got the new transformer yet, but it will be the same type as the existing one, only bigger and there is not a lot of room to do much. One good thing is that the transformer is an open frame type so it may be easier to do something.