Unconventional transformer use - UK Vintage Radio Repair and Restoration Discussion Forum

dougietamson · 11th Oct 2020, 4:07 pm

I have power transformer with multi-tap primaries for 117,150,165,220 and 240v input.
Today I'm getting 243v ac mains in the lab, the secondaries (no load) are measuring:
0-248v
3.4-0-3.4v
0-6.9v

I wouldn't expect much current from the HT, it has a 100mA fuse before the selenium rectifier ac input.

The 2 low voltage secondaries originally powered 5 valves, ef86, ef42, ecc81, el84 and em85.

So, what if I wanted a to build a valve tester by swapping the HT secondary to be the mains primary and using the old multi-tap primaries as multi-tap secondaries.

Here are the numbers I get out of it with no load:

117 = 111v
150 = 144v
165 = 159v
220 = 214v
240 = 234v

heater secondaries
6.8v and 3.3-0-3.3

Doug.

David Simpson · 11th Oct 2020, 4:38 pm

Hello Doug, I'm glad to see that you are considering building a DC Valve Tester.
Folk do use reversed transformers for projects, but take care regarding current & wattage ratings. I've used this method for a battery eliminator project, but being for 1920's TRF's the HT current consumtion was well less than 50mA.
Why are you using an old Selenium rectifier ? Plenty of modern Silicon ones available these days.
Have you got an ititial circuit diagram drawn up yet ? I've a few valve holders left, having just recently given away a couple of panels. Anything you are looking out for ?
If you are related to Jock Tamson, then you'll be related to me, eh ?

Regards, David

dougietamson · 11th Oct 2020, 5:43 pm

Hi David, aye, we're a' Jock Tamsons bairns.

Don't worry about the selenium rectifier, there is a fuse between the transformer HT out to the selenium bridge rectifier in the schematic, just so I could get a rough idea as to the limit of the HT current.

I think I lucked out with the winding ratio, pretty much 1:1 at each extreme of the pri sec HT windings.

I'll dig out my micrometer and measure the pri and sec windings.

I've had a look a few designs, Sussex one a bit overkill for me, the French lampemetre looks good (I lived and worked in France to 9 years, pas de souci!).
So, it seems to me, basically we need a power supply with a good range of voltages. For AC 5 and 6.3 maybe 12v too. For DC, 50 to 250 in 25v steps (here's where the multi-tap comes good and throw in some zeners on a rotary switch) to get us spot on. Then a negative DC bias 0-50v.
For the negative bias it would be nice to have stable regulation and a multi turn trimmer to fine tune when matching triodes.

The above transformer would be ok for up to 6v6 el84, I have some el34 and 6l6 which might cause a bit of sag.

For meters, those cheap £5 DT-830 jobs from ebay.

I'm good for bases, just noval and octal for now, keep the wiring short too. I have some loctal valves and sockets but would do a piggy back adapter for those.

What do you think?

Doug

Lucien Nunes · 11th Oct 2020, 5:49 pm

Beware possible lesser insulation between the original HT secondary that is now live with mains, and the heater winding. Also with the interwinding screen being in the wrong place, even if the insulation is good there would be still tend to be more capacitive leakage from the mains to the heater winding.

dougietamson · 11th Oct 2020, 5:58 pm

Thanks for the heads up, it's also 6 years older than me.

Radio Wrangler · 11th Oct 2020, 5:58 pm

Sort of sane. Most transformers will have had their turns ratios twiddled a little to allow for voltage drop in copper resistance. When you run them with an input other than the original primary, then these adjustments work against you and voltage ratios don't come out the same as you'd calculate.

Selenium rectifiers are horrible lossy things and one day will stink you out. Best avoided. Again, voltage drops will spoil the ratios you get from your transformer.

With bridge rectifiers and getting arty with the use of the previously primary windings, it's going to be a bit difficult to arrange for switching different voltages for anode and grid 2.

But then it's perfectly OK to have more than one transformer in a valve tester, so this could be your escape route.

I don't think it was long ago that someone on the forum was offering some ten-turn pots of various resistances.

David

dougietamson · 11th Oct 2020, 6:03 pm

Just had a closer look at the transformer, there are more primary taps giving:

36.6, 110, 136, 143, 158, 192, 211, 229 and 234v

dougietamson · 11th Oct 2020, 6:11 pm

Quote:

Originally Posted by Radio Wrangler

Sort of sane. Most transformers will have had their turns ratios twiddled a little to allow for voltage drop in copper resistance. When you run them with an input other than the original primary, then these adjustments work against you and voltage ratios don't come out the same as you'd calculate.

Selenium rectifiers are horrible lossy things and one day will stink you out. Best avoided. Again, voltage drops will spoil the ratios you get from your transformer.

With bridge rectifiers and getting arty with the use of the previously primary windings, it's going to be a bit difficult to arrange for switching different voltages for anode and grid 2.

But then it's perfectly OK to have more than one transformer in a valve tester, so this could be your escape route.

I don't think it was long ago that someone on the forum was offering some ten-turn pots of various resistances.

David

I'm not sure I kept the old selenium job, large bulky thing, are they toxic?

Electronpusher0 · 11th Oct 2020, 6:24 pm

Quote:

Originally Posted by dougietamson

the French lampemetre looks good (I lived and worked in France to 9 years, pas de souci!).

The Lampemetre article on the Web site is in French,
I posted a translation in this thread
https://www.vintage-radio.net/forum/...d.php?t=168699
Scroll down towards the end.

Peter

dougietamson · 11th Oct 2020, 6:35 pm

Quote:

Originally Posted by Electronpusher0

Quote:

Originally Posted by dougietamson

the French lampemetre looks good (I lived and worked in France to 9 years, pas de souci!).

The Lampemetre article on the Web site is in French,
I posted a translation in this thread
https://www.vintage-radio.net/forum/...d.php?t=168699
Scroll down towards the end.

Peter

Thanks Peter, my French is quite good, my wife got a job there, I fancied a break from my work (IT consultant). The kids went into French schools (nursery and primary school and were speaking French after 6 months, I did 2 years of classes learning the language at the age of 46!) It must have worked as I applied for a Job, tech part in English, HR part in French, got the job

dougietamson · 11th Oct 2020, 6:40 pm

Maybe I could run the transformer powering a SE EL84 amp with normal pri-sec, measure the current draw, monitor the temperature etc then flip the windings.

Doug

Electronpusher0 · 11th Oct 2020, 8:04 pm

Quote:

Originally Posted by dougietamson

Quote:

Originally Posted by Electronpusher0

Quote:

Originally Posted by dougietamson

the French lampemetre looks good (I lived and worked in France to 9 years, pas de souci!).

The Lampemetre article on the Web site is in French,
I posted a translation in this thread
https://www.vintage-radio.net/forum/...d.php?t=168699
Scroll down towards the end.

Peter

Thanks Peter, my French is quite good, my wife got a job there, I fancied a break from my work (IT consultant). The kids went into French schools (nursery and primary school and were speaking French after 6 months, I did 2 years of classes learning the language at the age of 46!) It must have worked as I applied for a Job, tech part in English, HR part in French, got the job

Is that what they call teaching my grandmother to suck eggs?

Oh well others may find the translation useful.

Peter

dougietamson · 11th Oct 2020, 8:33 pm

Quote:

Originally Posted by Electronpusher0

Quote:

Originally Posted by dougietamson

Quote:

Originally Posted by Electronpusher0

The Lampemetre article on the Web site is in French,
I posted a translation in this thread
https://www.vintage-radio.net/forum/...d.php?t=168699
Scroll down towards the end.

Peter

Thanks Peter, my French is quite good, my wife got a job there, I fancied a break from my work (IT consultant). The kids went into French schools (nursery and primary school and were speaking French after 6 months, I did 2 years of classes learning the language at the age of 46!) It must have worked as I applied for a Job, tech part in English, HR part in French, got the job

Is that what they call teaching my grandmother to suck eggs?

Oh well others may find the translation useful.

Peter

Not at all, having read the thread, you put a lot of hard work into getting a good translation.

I used to use google translate quite a lot when I was in the French office, just had to read it thru after to make sure it was what I meant to say.

David Simpson · 12th Oct 2020, 10:58 am

When I built my 1st DC V-T a few years back, it was initially very much a "suck it & see" project. Particularly for the power supplies. Each stage was carefully motitored for voltage, current, ripple & stability. So knew what to expect when finally fitting out metering for Va, Vs, Ia & Is. And, obviously, -ve Vg needed careful metering.
Dont be afraid of "re-purposing" other items of test equipment. For my original V-T I used a Marconi TF1041/C VTVM, and for the V-T I've just finished - an AVO8. Gives you accuracy & reliability.
Dump any ideas of using old Selenium rectifiers. Modern Silicon Bridges(eg. BY179) are cheapish & available.

Regards, David

Radio Wrangler · 12th Oct 2020, 2:45 pm

Selenium rectifiers have a well earned reputation for going off as stink bombs and make Rifa capacitors look mild.

Selenium is not terribly safe. Selenium is a teratogen, meaning it is linked to birth defects in quantities above the trace amounts the body needs.

As rectifiers, they aren't very efficient and aren't often very long lived. Best avoided.

David

dougietamson · 12th Oct 2020, 4:25 pm

Quote:

Originally Posted by Radio Wrangler

Selenium rectifiers have a well earned reputation for going off as stink bombs and make Rifa capacitors look mild.

Selenium is not terribly safe. Selenium is a teratogen, meaning it is linked to birth defects in quantities above the trace amounts the body needs.

As rectifiers, they aren't very efficient and aren't often very long lived. Best avoided.

David

It's in the bin already

I put the transformer in a single ended amp with output stage of 6V6GT and a couple of loctals for the pre-amp 7B7 pentode and 7C6(it has a nice 85u triode).
Ran it for an hour, the temp went up to 31c from an ambient of 20c with a
59mA draw on the ac HT and 780mA draw on the heaters.

Will be while before it cools down so will swap the pri/sec tomorrow and repeat.

Doug.

dougietamson · 13th Oct 2020, 11:54 am

Ran it in the se amp again today, this time for 2 hours, current draw was the same, temp went up to 33c.

The pri and sec HT windings use the same wire 0.4mm thickness, the wire used on the main filament with center tap is 0.9mm.

I could use the EF86 filament supply 0-6.8v for a bias voltage supply by way of a voltage double/treble/quadrupler.

I don't have anything more hungry than El34 and some Russian El36 equiv.

Doug.

11th Oct 2020, 4:07 pm	#1
dougietamson Hexode Join Date: Mar 2019 Location: Edinburgh, Scotland, UK. Posts: 341	Unconventional transformer use I have power transformer with multi-tap primaries for 117,150,165,220 and 240v input. Today I'm getting 243v ac mains in the lab, the secondaries (no load) are measuring: 0-248v 3.4-0-3.4v 0-6.9v I wouldn't expect much current from the HT, it has a 100mA fuse before the selenium rectifier ac input. The 2 low voltage secondaries originally powered 5 valves, ef86, ef42, ecc81, el84 and em85. So, what if I wanted a to build a valve tester by swapping the HT secondary to be the mains primary and using the old multi-tap primaries as multi-tap secondaries. Here are the numbers I get out of it with no load: 117 = 111v 150 = 144v 165 = 159v 220 = 214v 240 = 234v heater secondaries 6.8v and 3.3-0-3.3 Doug.

11th Oct 2020, 4:38 pm	#2
David Simpson Nonode Join Date: Mar 2008 Location: Aberdeen, UK. Posts: 2,852	Re: Is this sane? Hello Doug, I'm glad to see that you are considering building a DC Valve Tester. Folk do use reversed transformers for projects, but take care regarding current & wattage ratings. I've used this method for a battery eliminator project, but being for 1920's TRF's the HT current consumtion was well less than 50mA. Why are you using an old Selenium rectifier ? Plenty of modern Silicon ones available these days. Have you got an ititial circuit diagram drawn up yet ? I've a few valve holders left, having just recently given away a couple of panels. Anything you are looking out for ? If you are related to Jock Tamson, then you'll be related to me, eh ? Regards, David

11th Oct 2020, 5:43 pm	#3
dougietamson Hexode Join Date: Mar 2019 Location: Edinburgh, Scotland, UK. Posts: 341	Re: Is this sane? Hi David, aye, we're a' Jock Tamsons bairns. Don't worry about the selenium rectifier, there is a fuse between the transformer HT out to the selenium bridge rectifier in the schematic, just so I could get a rough idea as to the limit of the HT current. I think I lucked out with the winding ratio, pretty much 1:1 at each extreme of the pri sec HT windings. I'll dig out my micrometer and measure the pri and sec windings. I've had a look a few designs, Sussex one a bit overkill for me, the French lampemetre looks good (I lived and worked in France to 9 years, pas de souci!). So, it seems to me, basically we need a power supply with a good range of voltages. For AC 5 and 6.3 maybe 12v too. For DC, 50 to 250 in 25v steps (here's where the multi-tap comes good and throw in some zeners on a rotary switch) to get us spot on. Then a negative DC bias 0-50v. For the negative bias it would be nice to have stable regulation and a multi turn trimmer to fine tune when matching triodes. The above transformer would be ok for up to 6v6 el84, I have some el34 and 6l6 which might cause a bit of sag. For meters, those cheap £5 DT-830 jobs from ebay. I'm good for bases, just noval and octal for now, keep the wiring short too. I have some loctal valves and sockets but would do a piggy back adapter for those. What do you think? Doug

11th Oct 2020, 5:49 pm	#4
Lucien Nunes Rest in Peace Join Date: Apr 2008 Location: London, UK. Posts: 2,508	Re: Is this sane? Beware possible lesser insulation between the original HT secondary that is now live with mains, and the heater winding. Also with the interwinding screen being in the wrong place, even if the insulation is good there would be still tend to be more capacitive leakage from the mains to the heater winding.

11th Oct 2020, 5:58 pm	#5
dougietamson Hexode Join Date: Mar 2019 Location: Edinburgh, Scotland, UK. Posts: 341	Re: Is this sane? Thanks for the heads up, it's also 6 years older than me.

11th Oct 2020, 5:58 pm	#6
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,866	Re: Is this sane? Sort of sane. Most transformers will have had their turns ratios twiddled a little to allow for voltage drop in copper resistance. When you run them with an input other than the original primary, then these adjustments work against you and voltage ratios don't come out the same as you'd calculate. Selenium rectifiers are horrible lossy things and one day will stink you out. Best avoided. Again, voltage drops will spoil the ratios you get from your transformer. With bridge rectifiers and getting arty with the use of the previously primary windings, it's going to be a bit difficult to arrange for switching different voltages for anode and grid 2. But then it's perfectly OK to have more than one transformer in a valve tester, so this could be your escape route. I don't think it was long ago that someone on the forum was offering some ten-turn pots of various resistances. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

11th Oct 2020, 6:03 pm	#7
dougietamson Hexode Join Date: Mar 2019 Location: Edinburgh, Scotland, UK. Posts: 341	Re: Is this sane? Just had a closer look at the transformer, there are more primary taps giving: 36.6, 110, 136, 143, 158, 192, 211, 229 and 234v Attached Thumbnails Last edited by dougietamson; 11th Oct 2020 at 6:08 pm. Reason: add a photo

11th Oct 2020, 6:40 pm	#11
dougietamson Hexode Join Date: Mar 2019 Location: Edinburgh, Scotland, UK. Posts: 341	Re: Unconventional transformer use Maybe I could run the transformer powering a SE EL84 amp with normal pri-sec, measure the current draw, monitor the temperature etc then flip the windings. Doug

12th Oct 2020, 10:58 am	#14
David Simpson Nonode Join Date: Mar 2008 Location: Aberdeen, UK. Posts: 2,852	Re: Unconventional transformer use When I built my 1st DC V-T a few years back, it was initially very much a "suck it & see" project. Particularly for the power supplies. Each stage was carefully motitored for voltage, current, ripple & stability. So knew what to expect when finally fitting out metering for Va, Vs, Ia & Is. And, obviously, -ve Vg needed careful metering. Dont be afraid of "re-purposing" other items of test equipment. For my original V-T I used a Marconi TF1041/C VTVM, and for the V-T I've just finished - an AVO8. Gives you accuracy & reliability. Dump any ideas of using old Selenium rectifiers. Modern Silicon Bridges(eg. BY179) are cheapish & available. Regards, David

12th Oct 2020, 2:45 pm	#15
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,866	Re: Unconventional transformer use Selenium rectifiers have a well earned reputation for going off as stink bombs and make Rifa capacitors look mild. Selenium is not terribly safe. Selenium is a teratogen, meaning it is linked to birth defects in quantities above the trace amounts the body needs. As rectifiers, they aren't very efficient and aren't often very long lived. Best avoided. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

13th Oct 2020, 11:54 am	#17
dougietamson Hexode Join Date: Mar 2019 Location: Edinburgh, Scotland, UK. Posts: 341	Re: Unconventional transformer use Ran it in the se amp again today, this time for 2 hours, current draw was the same, temp went up to 33c. The pri and sec HT windings use the same wire 0.4mm thickness, the wire used on the main filament with center tap is 0.9mm. I could use the EF86 filament supply 0-6.8v for a bias voltage supply by way of a voltage double/treble/quadrupler. I don't have anything more hungry than El34 and some Russian El36 equiv. Doug.