Dropping mains voltage to mains transformer approx 20V

[high_vacuum_house]

Good evening,
I recently purchased an HP sampling voltmeter. This appears to work ok for a short time before it went silly but after cooling down would work ok for the same amount of time. The cause was heat related as leaving the cabinet and main board open the voltmeter would work for a long period.

The reason for the overheating was traced to the series pass transistors were dropping an extra 7V from the unregulated DC supply to get the plus and minus 15V rails regulated.

On running the voltmeter from a variac and taking measurements the mains supply needs to be around 220-230V for the unregulated DC voltage before the regulator circuitry corresponds to the manual. This tallies with the 230V tapping switch on the rear of the meter.

The power supply is fairly conventional with a filtered IEC socket, transformer, bridge rectifier, smoothing capacitors and then the Plus and minus 15 volt series pass voltage regulator transistors.

I measured my mains supply and it is surprisingly high at 248V Obviously too high for the voltmeter to be happy for long periods. What methods could I use to drop the mains voltage approx. 20V Dropper resistors in the case is out as the case is surprisingly airtight.

Many thanks,
Christopher Capener

[PJL]

Take a small 250V/20V transformer and wire it in series but out of phase with the primary winding. I use this technique for 220V European radios. It's called a bucking transformer.

[broadgage]

As above, and also a transformer with a 24 volt secondary should be fine if more available than 20 volts.

[bikerhifinut]

i had a thread on here recently describing how I made such a device for my power amps.
I wired mine as an Auto transformer rather than a bucking device. It just seemed a more elegant way but both methods do work.
if I can figure it out it must be easy!

Andy

[high_vacuum_house]

Many thanks for the replies,
I shall try and find a small 20-24V mains transformer to fit in the space available inside.

Christopher Capener

[TrevorG3VLF]

Is the voltmeter made for 110V 60Hz input?

[Boater Sam]

A very long extension cable?

[Skywave]

The above suggestions incorporating an external transformer are sensible and will solve the problem. One possible snag to them is that the additional transformer will be external to the item, which apart from a 'clumsiness' aspect also includes the possibility of it accidentally not being used or even lost / mislaid. That can be avoided, of course, by a 'hard-wiring' style, but that brings us back to the 'clumsiness' aspect.
Is it possible to re-engineer the item by the addition of a heat-sink at the rear of the case and to relocate the series-pass transistors to that sink?

Another alternative is to broaden your view of this and consider your mains supply voltage: you stated 248 v. That figure causes me to wonder about the rest of the equipment you have there - running on that voltage. Perhaps a suitably-sized transformer is called for, to reduce the mains supply to a nominal 230-v. for all of that equipment? It might be an opportune moment to consider the fitment of an isolating transformer.

Al.

[Herald1360]

The bucking transformer can be quite small. It only needs to be rated for the mains supply current of the instrument at its secondary voltage.

eg a 100mA mains (24VA) instrument with a 20V drop needed only requires a 2-3VA bucking transformer.

[mhennessy]

248V is in spec for UK mains (230V +10%, -6%; 216-253V).

Without knowing the model number, I'd be surprised if this high quality HP device would be unhappy with this - if that was the case, the designers would have included taps on the transformer to adjust for the expected range of AC voltages. Remember - it used to be 240V +/-6% (226-254V) in the UK.

Following that thought-process through, that suggests to me that there is a thermal fault with the instrument. It could be that the series pass transistors could do with their thermal grease replacing and the screw connections re-torquing. It's also possible that one of them is dying - I've certainly had old power transistors play up when hot (current gain falls away). Could even be capacitors...

The voltages prior to the regulators will be nominal at best, and the designers would have taking into account the full range of expected voltages.

Give us the model number, and perhaps David (Radio Wrangler) might be able to comment further. But certainly, that's where I would go first.

Mark