Paralleling windings on a Mains Transformer

[Old Sparky]

In order to double up the current capability of a LT secondary winding, which is 5-05v CT@150mA approx (it's part of an old Murphy M281PSU intended for use with the '96' range of Valves, and I want to use it in a radio using the '91/92' range, whose filaments draw 50mA each, as opposed to the 25mA each of the '96' series) I thought of simply joining the two outer ends of the winding in parallel, until I sketcked it out, then I realised that doing so would effectively short-circuit the winding. The centre tap is not easily accesible in order to modify the connections, so I presume that I can't go down this route, right?

[AlanBeckett]

This is one for Ed, but I fear you're correct.
On the other hand small Heater Transformers giving 5 Volts at a couple of Amps are pretty readily available.

Alan

[pwdrive]

Whatever transformer you use don't forget to rectify the output.

Lawrence.

[tiv48]

I think you have to check if the two windings are in phase. (probably not)
you will need to separate the centre tap and wire both windings in parallel.
Rob T

[Refugee]

I seperated the ends of a winding on a slightly larger transformer in just this way.
It lasted about a month then went bang.

[Eddystonepete]

Hi Sparky, have you considered trying the existing transformer with say a lamp load to see if the voltage holds up and the tranny doesn't overheat? Murphy was probably generous with ratings anyway. Parallel windings are ok but any voltage imbalance will give losses ie heat.

Peter

[Old Sparky]

What I found, some time ago, was that, with a full wave rectified o/p followed by a series pre-set resistor and a couple of 1N400x diodes in series across the o/p to ensure that the voltage applied to the filaments couldn't rise above 1.4v, that, if I set the 30 ohm pot. to give close to 1.4v the radio operated OK, but the load on the winding was too great, even after disconnecting the 12v LES bulb connected across the LT winding (part of the original Murphy Design) After all the 5 filaments in parallel draw 250mA, which is double what 5 filaments in a set fitted with '96' range valves consumes.

[Nick]

If you can get to the centre tap and split the secondary into two 0-5VAC windings, then you can parallel them up so long as you are really careful that they are in phase - a bit of a write-up on this can be found at the Elliot Sound Product site - http://sound.westhost.com/xfmr2.htm (good site anyway).

Obviously, you will need to change the full bridge rectification from 2 diodes to 4 as there is now no centre 0V reference.

If you can't get to the centre tap, then all bets are off I'm afraid - whilst Murphy may well have provided some headroom on their transformers, that would be for a reason, and anyway its not likely to be as large as 100% - further, if you use that headroom, then your new application has a much reduced safety factor.

I wouldn't go there...

[jimmc101]

At the risk of being accused of heresy, how about retaining the full wave rectifier and replacing the linear regulator with a a buck converter to gain the extra current?
O.K. it's going to generate interference but something like this http://www.ebay.co.uk/itm/Mini-DC-DC-Step-Down-Buck-Power-Supply-Module-Adjustable-Converter-MP1584-/251588582445?pt=UK_BOI_Electrical_Test_Measurement_Equipment_ET&hash=item3a93d91c2d
based on the MP1584 can be pushed to operate at 1.5MHz and so might be good enough if filtered & screened well.

There are chips available with much higher switching frequencies but a quick look didn't find any complete modules.

Jim

[Nick]

An SMPS is not a bad plan - you could hide it as its small and as its switching mostly out of the MW band, likely to be fairly quiet. Maybe make a small sealed FR4 soldered box with feedthrough capacitors on input and output power lines and the FR4 connected to GND... Avionics kit and mobile phones are full of SMPS units - at the price of that auction, I'd give it a go... Note that significant noise is reflected down Vin (because of the way SMPSs work), so that has to be filtered too...

I use a MAX1771 based boost converter for small valve amps and they're fine -switching is at about 100khz for those so I do a lot of filtering... I have used the Murata BNX002-1, which even though its officially rated at 50V, seems perfectly happy.... I only used these as I have a box of them...