Dougs trioid transformers

[jeff]

I was looking at Dougs trioid transformer schematic. He is using a step down tranny as a step up tranny and a 12VAC walwart. But in the schematic it looks like the two 9V taps are wired in parallel not in series.
 Is it ok that each 9V tap is seeing 12V?

[EL34]

You have to think about the wiring backwards if this were used the normal way

The primary is on the left in the image
The primary is able to be wired for 240 or 120

So you feed 12 volts ac into the secondary and you are getting your B+ out of the primary

The image below is wired correctly for step up and my 12 volt AC wall warts

[jeff]

Thanks, I think I get what your saying. And your schematic is correct, no doubt. But what's in question is the transformer itself.

I get that you can take a 120V:9V tranny and put 120V in and get 9V out or reverse it and put 9V in and get 120V out.
 
But my question is: is it ok to use a 12V walwart with a 9V:120V stepup tranny? if I'm reading this schematic right your putting 12V into each 9V tap.


If the right side was in series then you'd be putting 12V across two 9V tap, each 9V tap would be getting 6V, that's 3V UNDER it's rating.
But since the taps on the right are wired in parallel each 9V tap is getting 12V, that's 3V OVER it's rating.
I imagine you could put 3V less than a tranny is rated for and still be ok, but is putting 3V MORE than the tranny is rated for OK?

also wouldn't you be getting much more than 240V out?
 9V:120V = 12V:160V,
 160V X 2 =320V,
 320VAC X 1.4(ss rect) = 448VDC

[EL34]

They are not 9 volt taps, it's just the secondary winding
So you feed whatever AC into the secondaty and out comes a higher AC from the primary


And the primary is in series so you get way higher than 120vac


My tube pedal projects shows I had a B+ of 284 VDC
http://el34world.com/projects/tube_box_4.htm

[jeff]

If you take a 120V:9V and feed it(backwards) 12V do you get 160V out?
And if you take a 120V:9V and feed it 6V do you get 80V out?

[jeff]

So if the 9V side is in parallel and the 120V side is in series and you give it 12V  each 9V"primary" gets 12V and you'll get(160+160) 320Vac out.
320Vac X 1.4(rect) = 448Vdc

But if the 9V side is in series and the 120V side are in seires and you give it 12V each 9V"primary" gets 6V and you'll get (80+80) 160Vac out.
 160Vac X 1.4 = 224Vdc.

[goldstache]

I use this power supply in a bunch of stuff.  I get about 280VDC (unloaded) once it hits the reservoir cap post rectification.  My supply is rated at 12VAC (14VAC) unloaded.  I prefer using a Lm317 to obtain 6.3VDC for parallel heaters.  Once everything is up and running depending on stages, the 12VDC series heater rail sags below 12.6VDC (11.58). I like to run it as close to heater spec for tube life. 

The tranformer to my understanding, is a turns ratio so Voltage output is proportional to input voltage.
Hope this real scenario helps

[Willabe]

I like to run it as close to heater spec for tube life.

Spec is +/- 10% of 6.3 on tube charts. Not necessary to nail 6.3 exactly.


                          Brad     

[jeff]

Ok I guess I don't get it. I'm not trying to be arguementitive, really, I'm just missing something an am trying to understand.

If you have a 120V:9V transformer and hook up 9V to the secondary will you get 120V on the primary or will it be less that 120V due to some reason(losses, inefficency, etc)? 

In other words is a 120V:9V xfmr a 9V:120V. Or when used backwards you don't get the same voltage ratio as going forward?

[PRR]

In a strict sense, you should not put 12VAC into a 9VAC winding. If the applied voltage is much higher than the rated voltage, the transformer saturates, over-heats.

However some small transformers have ample margin and/or just don't get hot easily. Doug has been selling that set-up for years and we know it works OK.

As for the general plan of 120:12V to 12:120V-- with small iron you may lose 20% in each core. So the loss of two cores could be 40%. What you get out may be considerably less than you put in. OTOH, if your load is small, losses are less, and maybe you wind up with enough juice to do what you want.

Try it and see. If you try it with other transformers, be alert for overheating.

[kagliostro]

About overheating, can the fact that tose are toroidal transformer have a positive effect on avoiding it ?


K

[EL34]

I did not come up with the Toroidal power supply idea

Electro Harmonix has been using them in pedal for many years.
I got the idea from them.

The 12vac wall warts I sell, the oddball female power jack and the tube pedal chassis all are from New Sensor/Electro Harmonix

I have sold hundreds of the toroidal trannys over the years.


These are real world voltage measurements, not calculated

[kagliostro]

the oddball female power jack

I was wondering which kind of female jack you use

.... it is an old Speaker Jach, who knows what I thought was

Franco

[EL34]

The wall warts I stock have the the oddball male plug on them


And so I carry the matching oddball female jacks so you don't have to cut off the oddball plug and replace it with a more common plug

[jeff]

Ok Thanks,
 Please don't get me wrong. I believe 100% that those are the real world values, and trust you 100%.
My questioning is Not me saying "that's not what I get when I calculate it, so your real world values must be wrong". but rather "that's not what I get when I calculate so I Know there must be something wrong with the way I Am Calculating" And trying to figure out the error in My thinking.
I hope you understand my intent, I accept that I Am wrong, I'm only questioning to Understand Why I am wrong.


Thank you for your time and patience
Jeff

[sluckey]

I accept that I Am wrong, I'm only questioning to Understand Why I am wrong.

There's nothing wrong with your calculations. But you have not considered the internal losses of the transformer or the loading effects of the external circuit. Remember, this is a tiny 1.5" x 1.5" x .75" transformer. Even the loading of rectifier diodes and filter caps may be a factor, not to mention the 1-2mA load of each triode. Sounds like a light load, but when connected to this tiny transformer, that light load becomes significant.

[jeff]

Something just dawned on me and I need to ask this question because I think I see the error of my ways here.
I was looking at this as two seperate 120V:9V transformers in one package but I think that assumption may be totally wrong. Now that I've spent some time thinking about it, would it be more correct to think of it as one transformer whose primary can be connected for 120V or 240V operation depending on how it's wired with to seperate 9V taps.


say we connect it for step down operation

If you only connected just one of the 120V primaries to 120V and didn't use the other 120V primary at all(instead of connecting them both in parallel):

would you get
- voltage on only one 9Vtap(as if it were 2 xfmrs in one package and the 2nd xfmr is not being used)
 or
- voltage on both 9V taps(but less avalible current draw from each because we're only using half the primary)

[kagliostro]

If the transformer has 2 x 115v primary windings and 2 x 9v secondary windings and you connect only one 115v primary winding

you obtain a 115v secondary and 2 x 9v secondary

but that at the cost of 1/2 total VA disposable

so if the transformer is rated for 20VA you can consider it as a 10VA, the sum of disposable current on the 2 x 9v secondary and in the 115v secondary is equal to 10VA to which you must detract the intrinsec lost of the transformer


K

[jeff]

OK thanks, I think that's where I went wrong, thinking of it as 2 seperate 120V:9V Xfmrs.
I'm gonna take some time to let that sink in and think about that before I respond this time but I think I get it now.
Thank you all very much(everybody)!

[jeff]

But you have not considered the internal losses of the transformer or the loading effects of the external circuit. Remember, this is a tiny 1.5" x 1.5" x .75" transformer. Even the loading of rectifier diodes and filter caps may be a factor, not to mention the 1-2mA load of each triode. Sounds like a light load, but when connected to this tiny transformer, that light load becomes significant.

Let me see if I understand this correctly
 
A 120V:12V transformer may not be a 12V:120V under load because the load in now on the primary and the "under load" ratings only apply when the load is on the secondary?
 
Say you take a 12V:120V 450mA transformer and wired it correctly-120V on primary. With no load you may read more than 12V, but with a 450mA load connected to the secondary the voltage drops and the 120V:12V at 450mA ratings hold true.
 
BUT>>> when you wire it backwards and put 12V on the secondary and put a load on the primary that 120V:12V rating no loanger applies because now the primary is loaded and that's not what the 120V:12V rating was designed for.

Seems to make sense the more I think about it. extreme example but I couldn't just take 45 transformers and wire them back to back to back to back to back' etc. and expect no loss. 45 transformers can't be equal to one transformer's performance.
 
So under load the first Xfmr may give a 120V:12V ratio then the next may give a 12V:100V ratio then the next may give a 100V:5V then the next 5V: 83V etc.etc. and that 120V:12V ratio doesn't hold true both ways

Does that sound right?

[sluckey]

But you have not considered the internal losses of the transformer or the loading effects of the external circuit. Remember, this is a tiny 1.5" x 1.5" x .75" transformer. Even the loading of rectifier diodes and filter caps may be a factor, not to mention the 1-2mA load of each triode. Sounds like a light load, but when connected to this tiny transformer, that light load becomes significant.

Let me see if I understand this correctly
 
A 120V:12V transformer may not be a 12V:120V under load because the load in now on the primary and the "under load" ratings only apply when the load is on the secondary?

How do you get from what I said to this understanding? That's a big leap IMO.

I don't think you will be satisfied with anything that's said. I think you should throw all the calculations out the window in this case and just accept the numbers that Doug and goldstache have provided. Or buy Doug's wallwart and tiny transformer and experiment with them.

[EL34]

LOL, What Steve said


Buy the parts, hook them up and be done with it
During that time you can take real world measurements.


Calculations are fine, but scientist do experiments after calculating to get real world results

[jeff]

I don't think you will be satisfied with anything that's said. I think you should throw all the calculations out the window in this case and just accept the numbers

Understand what you're saying. I totally do acept those numbers. I'm sure if when I buy the parts and experiment I'm 100% sure I'll get the exact same readings you got, ...but I still won't understand why I'm getting those readings. I get that my calculations are garbage, I do. 
I believe that this transformer in this application does in fact work, that's no longer in question.
 (I only mentioned dougs transformer specifically when I started this post because to me it looked like there was 12V on a tap rated for only 9V and that lead me to questions about how transformers work in general when they're connected backwards. I totally trust you guys, and believe you guys when you tell me it's ok to put 12V on the 9V tap I just don't understand it and that caused me to question my misconceptions about how a xformer works when wired backwards).

The last post was trying to figure why it makes a difference if the transformer is teeny weeny or not.
Let me ask this.
If you have a 10mA 120V:12V xformer  it will read 12V when the load draws 10mA
And if you have a 100A 120V:12V xfmr it will read 12V when the load draws 100A
But, and this is what I trying to understand, would it be correct to say when unloaded both secondaries won't read 12V and in fact the teeny weeny one will read much lower(unloaded) than the 100A beast(unloaded).
So unloaded one may read 120V:12.6V and one may be 120V:17V but loaded with thier appropreite loads each should read 120V:12V(is that part right?)

and if that's true does that significantly change how the teeny weeny xformer operates vs how the 1mA beast will operate when wired backwards? Is that what you meant by mentioning it was tiny? That's how I took it. I'm trying to understand

[PRR]

> how transformers work in general when they're connected backwards.

This is a complex question and I should be in bed.

The (typical) 20% regulation works forward or backward. So 20%*20% is 40% regulation and a very significant drop.

The Turns-Ratio usually allows for regulation. A "120:12" is probably wound for 120:14, with a few ohms resistance, so that under full load it hits its 12V spec. When you work backward, this step-up becomes a step-down.

True, a 120:12 - 12:120 rig should self-correct at no-load, but at full load it will sag bad. Because of resistance, and because you have un-done the winder's clever step-up.

> why it makes a difference if the transformer is teeny weeny

This requires an appreciation of Dimensional Analysis, and it is really too late to go into. However.....

Consider two transformers:

#1 -- 1"x1"x1"
#2 -- 10"x10"x10"

Surface area:

#1 -- 6 square inches
#2 -- 600 square inches

Cubic volume:

#1 -- 1 cubic inch
#2 -- 1,000 cubic inches

Power capacity goes roughly by Weight, and weight goes by cubic capacity. #2 should be good for 1,000 times as many Watts.

However we have losses which means Heat and we shed heat with Surface Area. #2 can only dissipate 100 times the heat of #1.

Finally: transformers always cost too much, so we always sharp-pencil then as small (cheap) as possible.

#1 has plenty of surface area, won't over-heat, even if we use extra thin wire and allow large losses.

#2 will over-heat easily, we have to size the wire to keep losses down just so it does not smoke.

Here's a graph from a book which goes way-too-deep into details. There's a lot going on here, print it and stare at it a while.

Across the bottom is transformer capacity in VA (Watts, near enuff).

Up the side is Regulation, 1% (super stiff) to 100% super-saggy).

Diagonal lines are Heat Rise. 50 deg C is a good number for insulation breakdown. We can pay much more for higher-temp stuff, slightly less for wimpy stuff.

I have drawn a "COOK" line near 50 deg C. Pick your VA/watts on the bottom, read up to 50 deg C, then read over to the left. This would be the Regulation for a transformer designed to run that hot.

But at 10VA, the Regulation "should" be 50%. This is very saggy! Basically we can't make a small transformer cook itself (surface/volume ratio). Design must be on other basis.

Often we would like "good" regulation, maybe 10% ? But below 100VA we have to add material to get good regulation, even though the heat is fine. Much below 100VA it gets silly and we find that a 20% regulation is a better buy, even though we may have to up-rate some of the parts of the load for the uncertain voltage. Down around 10VA we either buy much bigger than we need or we accept "terrible" regulation. Doorbell transformers can be 40% high when un-loaded... the un-connected doorbell does not care, and sizing the door button switch for 40% excess voltage is trivial. With that we also have a part that CAN be shorted more-or-less forever-- it sags so fast that the total heat is less than it can comfortably dissipate even with crap insulation.