Transformer (step down) -> transformer (step up) tube power supply

[brett]

Back when, I built a firefly tube amp (about 1 watt?).  Not knowing much, and being a cheapskate, I used a wall-wart transformer to step my Australian 240V down to 9v, then fed that into another wall-wart transformer to step the voltage back up.  The second might have been a 12V to 240 V transformer, giving about 180V to rectify for ??300V DC.
I now realise that I was naive, not considering some potential problems.
So my questions are:
1.  Is there excess inrush current when switched on?  My little fuse (100mA fast blo) didn't ever blow.
2.  Would it be easier or safer to generate a "near" 9V sine wave from a 12V DC wall-wart and step that up?  ie avoid dealing with a 240V mains transformer (under stress?) in the cabinet?
3.  Or...Use a 12V AC wall wart, and step that up?  Is a 12V AC wall-wart likely to be regulated or fused?  Adding useful safety to the design?
4.  Other than some higher than normal losses due to inefficient winding, is there another problem with stepping up using a step diwn transformer?

Thanks for your help!
So much help from you all over the years.

[Rob Strand]

It's related to the magnetizing current.   All transformers have a magnetizing current even when there is no load.   For small transformers the magnetizing current can be a significant fraction of the rating.  When operating a transformer in reverse with the rated voltage applied to the secondary it can be larger than the transformer can handle.

There's a lot of info posted on this thread, which gives some tricks how to reduce the magnetizing current,
https://www.diystompboxes.com/smfforum/index.php?topic=121216.0

If you want to know what is going on you need to at least measure the secondary currents with and without the HV load.

Recently I posted some easier to understand info  I'll try to find the post.
Here,
https://www.diystompboxes.com/smfforum/index.php?topic=129257.0

[brett]

Thanks for your helpful reply.
I should have mentioned that this is only for a preamp.  Using a 12AX7.
My plan is to use a 12V 1A wall wart, put a current-limiting resistor (10ohms) before the step-up and do conventional rectification and filtering at the output (giving likely somewhere between 120V and 180V AC and 150-220V DC).
The B+ runs at around 3mA, so the wall-wart runs at a theoretical minimum of 60mA and possibly 120mA given losses (guess). 
But that's ok for a 1 A wallwart, even if it's also running the heaters.
Thanks again.

[PRR]

> Is there excess inrush current when switched on?

What is "excess"? Does the house burn down? Does the transformer burn-up?

People (especially on another, hi-fi, forum) are IMHO much too "nice" to their wall-outlets. Utility power can take huge overloads for short times. My well-pump settles down to 11 Amperes running steady, but the starting surge is 44 Amps. Has been since it was installed mid-1980s. You find similar motors and other LARGE loads in century-old factories, bang-started tens of thousands of times, often unmaintained and un-failed.

My well-pump is a minor annoyance because the lights dimmed, although "60W" LED bulb hardly stutter. (The 1W night-lights can flash badly.) But a 12VA transformer isn't dimming your lights.

> 12V 1A wall wart, put a current-limiting resistor (10ohms)

A waste of 10 Ohms. There's 10 or 20 Ohms already in the windings.

If your heaters are hotter than your transformer, just play it.

[Rob Strand]

The B+ runs at around 3mA, so the wall-wart runs at a theoretical minimum of 60mA and possibly 120mA given losses (guess).
But that's ok for a 1 A wallwart, even if it's also running the heaters.

What you will find is the actual secondary current on the reversed transformer is much higher than those estimates.  The magnetization can easily be 200mA or perhaps a lot more depending on the reversed transformer.    The magnetization current is not just at power up, it's continuous.

Adding a series resistance can help a bit.   When the voltage across the secondary of the reversed transformer is reduced the magnetizing current reduce.  You can make great reductions in magnetizing current when the secondary voltage at the transformer is 0.5 to 0.7 the rated voltage.    The resistor causes a small voltage drop, that reduces the voltage to the secondary and it help shave off some current.  You might find you will need a fairly large value.   The side effects of off that is the HV side voltage reversed transformer will drop.   You can't actually have low current and maintain the voltage.   

A better way to shave off magnetizing current is to feed say 12VAC into an 18V reversed transformer.   Again the output on the HV side will be reduce, exactly the same way is reduced with a resistor.   The advantage here is you aren't making more heat in the dropping resistor, and the voltage probably won't drop as much under load.

The idea behind the resonate cap is to shave off magnetizing current without trading off output voltage.

As mentioned in the second thread I linked using a reversed *toroid* transformer will guarantee a low magnetizing current but conventional EI cores are quite variable.

[merlinb]

Why are you even asking the question? Are you worried something is going to burn out? Transformers are very hardy...

[brett]

Thanks all.  As I mentioned at the start, I ran a couple of 240 to 12 VAC transformers back-to-back and from memory got enough out to get 305V of highly regulated DC.
So I'm sure that it works, and that case it supplied a much higher current than I require now.
Perhaps more importantly than anything else, I'll make lots of measurements and observations in anything I build now.

[Ben N]

Does the house burn down? Does the transformer burn-up?

Why do houses and transformers burn in opposite directions?

[antonis]

If you really worry about inrush current (where you shouldn't..), you can implement some kind of delay/soft start circuit..
(but BEWARE of parts connected to mains..!!!)

[mac]

I use two old dichroic 220v-12v irons to experiment with tubes on the breadboard, no gloves.
I connect the heaters to 12v AC and sometimes 100 ohms resistors from both extremes to ground to reduce AC noise. These are your "fuses/shunt" if any "inrush current".
The heaters load the first iron a bit so there is less than 220v AC at the output of the second one, no big deal.
My only concern is two irons can make a lot of noise, you should rotate them.

mac

PS: I'm alive, and I never have to call the firemen.

PS2: loading the 12v AC --> Eddie's variac --> brown sound    

[Rob Strand]

The issue isn't inrush.  That's an assumption the OP made.

The high secondary current (ie. LV winding) is there all the time.   
The current is larger than the fuse rating so it blows.

Measure the AC current and it will be obvious what is going on.

[PRR]

A better way to shave off magnetizing current is to feed say 12VAC into an 18V reversed transformer.

Brett says he was going: "240V down to 9v, then ...<snip-snip> ....might have been a 12V to 240 V transformer".

So 9V into a 12V winding.

Is 12 to 18 more mojo than 9 to 12? (OK, 0.125 different ratio.)

If the heat of the 12VA transformer amounts to 25% loss, then 24/7 for a year is about $7 on your electric bill (say a half buck billed monthly). Take a decade to justify the cost of a better scheme.

The current is larger than the fuse rating so it blows.

My little fuse (100mA fast blo) didn't ever blow.

I think this is over-thought.

[Rob Strand]

Brett says he was going: "240V down to 9v, then ...<snip-snip> ....might have been a 12V to 240 V transformer".
...
I think this is over-thought.

It's actually not clear what he's done, asking for advice on (better to use 9V?), or intends (later saying 12V 1A).

What I can say is some 12V EI transformers even running from 9V will pull more current than their rating (whereas others might be under the limit with 12V into 12V).    That's particularly true for transformer smaller than 2VA perhaps stretching upto 4VA.    Large transformers are less likely to have problems.

The 9v in /12V reverse = 0.75  is a little worse off than 12V in / 18V reverse = 0.67.    What happens is the lower voltage naturally pulls less current but the inductance *rises* so you make double gains in reducing the current.   The inductance maxes out at around 0.5 to 0.67 of the rated voltage.   However crappy small transformers with large effective air-gaps don't behave like that.  They tend to pull heaps of current and aren't suitable for reversing.  (Those same transformers might run close to 40C above ambient with no load in the forward direction.)

How much input power isn't the problem.  The problem is frying the reversed transformer.    If the input transformer rating is the same or lower than to the reversed transformer then that can fry too.

[anotherjim]

I've only made a couple of projects with down-up transformers but have not noticed any heating issues.
One is a 1U F2B rack clone with a 6v toroid driving a laydown 9v/220v EI. The 6v is also feeding bridge rectified & regulated DC for the two heaters. Get enough raw DC for a 7806.  The backward with a bridge rectifier, the HT is about 180v.

The other project is a small ECL86-based amp. An EI PT supplies 6vAC for the heaters but the backward is a bit different. I used an old Eagle brand 220-240v battery eliminator transformer. This has a multitap secondary so I was able to fit a switch to pick the step-up ratio. Finding usable taps wasn't so straightforward going from the 0 end of the secondary, but there were enough taps to take voltage between to suit the 6v input. I can't remember now what I went with, but I got 2 usable settings and fitted a switch for them. This happens to be the lowest hum class-A I've ever heard with AC heaters and no elevated DC level tricks.

For the input transformers, I don't think 9v secondaries are of much value. You have to provide 6v or 12v for heaters. The 9v will do well backward from a 6v input.

[antonis]

The issue isn't inrush.  That's an assumption the OP made.

We can't say anything in the absence of additional data, like reservoir capacitor(s) size..