Checking unknown PSU transformer mains voltage (Advantest R6551)

[vutt]

Disclaimer: Since we are dealing potentially hazardous mains voltage levels here I'll keep it for time being as theoretical discussion apart from some multimeter measuring action.

I have old Advantest R6551 DMM with voltage selector hard soldered for 100V. According to spec in the back 90 - 110V. In my region (230V) I have only access to standard US-EU transformers 230->115V. Unit is working but 115V is +5v out of my unit currently set spec - attached pic1 upper corner. Don't want to buy another "humming" transformer. Better to convert unit PSU and learn something while doing it.

While I know for the fact even from this forum that this DMM can be converted to 220-240V my goal with this thread is more like how one could investigate PSU mains input conversion possibility. And more importantly how to double triple check final solution without actually powering unit up.

Theory.
I'm aware of basic transformer working principles. I bought cheap Chinese variac/auto transformer in order to verify my knowledge. Fine tuned main input voltage and made some secondary winding voltage measurements going to DMM analog portion. See pic2 - clear correlation between transformer input and output. It looks like up to 0.85V overvoltage (110->115) on 14v AC rail did not damage anything.
I also measured AC voltages going to bridge rectifier and V showed similar correlation.

DMM PCB markings clearly indicate primary windings jumper bridges W0(1..3) for 120,220 or 240 volts. With this in mind I made some DC resistance measurements between G-point  and all three primary windings entry points AC120 - AC240 (Pic1). DC resistance numbers kind of support idea that I should bridge W03/AC240 (Point A on pic) and cut PCB connection on point B.

Q: What kind of methods I have in order to verify my solution? I'm aware of function generator method to verify transformer ratios, but it involves disconnecting transformer completely. In my case I can desolder secondary analog board wires, but I cant disconnect transformer from bridge rectifier width out desoldering entire transformer.
Could I use auto transformer, bring slowly mains voltage up from 0 to 230 and monitor closely secondary windings output?
I have also new fancy LCR meter (ET4410) but I'm still learning basics with it. That's why I bought it. Could it help me to have extra test use case for my solution?

[bob91343]

If you think you have it wired correctly, bring up the power gradually with the variac while measuring the input current.  If the current doesn't get too large, you are probably okay.

Miswiring can have two effects.  One is that two windings in parallel have different numbers of turns and thus will cause excessive currents with very low input voltages.  The other is that the output voltage will not be correct - either too large, possibly causing damage, or too small, possibly not enough for normal operation.

[edpalmer42]

This is an unusual configuration for selecting mains voltage.  Typically, there are two windings.  Put them in series for 220/240 V and in parallel for 100/120 V.  The 220 and 100 V settings use taps on one of the windings.  This requires the movement of two jumpers to change the voltage.  Your unit doesn't look anything like that.  It looks like there are seperate 120V and 240V windings with taps for 100 and 220V.

The first thing you could try is to look around for similar low-power devices from Advantest.  It's reasonable to assume that they might have used a similar arrangement in other devices.  If you can find a service manual, it might give you an explanation.

Other measurements that you could make would be to measure resistances as follows:  100 - 120 V, 220 - 240 V, and 120 - 240 V.  Do those measurements agree my assumption above or suggest something different?

[bdunham7]

From you measurements it does appear that your theory is correct, but cutting a trace doesn't seem like the way they would have intended to do it, so I'd be wary of a trap.  YOu could further test by powering it up as-is with 100VAC and then measuring from the AC240 point to neutral--you should see about 240VAC if the windings are actually all in series.

Can you move those wires out of the way and provide a clear shot of the top side of the board?

[vutt]

@bdunham7

Attached "xray" pic and one pic from random Chinese forum with "evidence" how 220v jumper looks.
I measured ~230V between "N" and W01/AC240 transformer point with current 100V powering setting. Not exactly 240 but close enough I guess.

[vutt]

Other measurements that you could make would be to measure resistances as follows:  100 - 120 V, 220 - 240 V, and 120 - 240 V.  Do those measurements agree my assumption above or suggest something different?

  20 Ω for 100 - 120 V
  67 Ω for 220 - 240 V
393 Ω for 120 - 240 V

[bdunham7]

So it looks like there is one 240V winding (of varying wire gauge) or windings in series, however you like to think about it, with taps for 100/120/220 volts.  However much cutting a trace in this situation seems odd to me, I can find no flaw in your reasoning.  I take it the trace going down from the black wire goes to the power switch and then back up to the jumper/tap area?

A manual would be nice, but they seem to be totally unavailable.

[edpalmer42]

So it looks like there is one 240V winding (of varying wire gauge) or windings in series, however you like to think about it, with taps for 100/120/220 volts.  However much cutting a trace in this situation seems odd to me, I can find no flaw in your reasoning.

I've seen series windings like this occasionally.  I agree that cutting a trace is very odd.  They obviously are tightly focussed on the Japanese market.

[vutt]

Well good news is that I managed to do it with my eyebrows still in place. Even my apartment building still has the lights   

Pic below is reference test with 100V sharp mains voltage. Power consumption 0.086A
Amazingly DMM did not act crazy when I started to add voltage from 0V. Just at one point according to spec it turned on and booted up normally.
Todays mains output flavour for me was at 233V mark. I decided to jump AC240 bridge.
After conversion power consumption was around 0.033A. Secondary AC voltages measured however little bit below compared to reference voltages seen on pic - 3.6V and 13.1.
Don't worry I checked fuse before conversion. To my surprise it had 800mA F fuse.  Clearly too fast and too big now. Needs to be replaced...

I have to admit it felt very medieval cutting that trace.

All in all thanks everyone who helped me with this little adventure!

Bonus emotion - Finally I managed to convince myself that having 3+ DMM-s in household is not waste of money...

[edpalmer42]

Congratilations Vutt!    

Yes, a collection of DMMs looks odd at first glance, but there are occasions where you need them all!