Transformers in parallel?

[Colinaps]

I should check carefully what happens when any one diode fails, in case you inadvertently make a voltage doubler, or worse.
Cheers,
Colin.

[ms660]

Originally Posted by kalee20

whether the OP has his several tens of thousands of microfarads, or a hundred million farads. (This latter extreme would of course take a lot longer to reach steady-state, but once there, the transformer wouldn't notice the difference).

But the diodes might...

Lawrence.

[kalee20]

No, the diodes wouldn't see any significant difference either, under steady-state conditions.

It's only during the period of getting to steady-state - or during changes to steady-state (such as mains brown-outs or surges) , that really huge reservoir capacitors have an impact. Hence thermionic diodes having a maximum C limit.

[ms660]

Originally Posted by kalee20

No, the diodes wouldn't see any significant difference either, under steady-state conditions.

It's only during the period of getting to steady-state - or during changes to steady-state (such as mains brown-outs or surges) , that really huge reservoir capacitors have an impact. Hence thermionic diodes having a maximum C limit.

I was meaning during the initial charge period...

Lawrence.

[ms660]

Screen shot of a simulation for all of the secondary windings in parallel, keeping in mind that I'm not a transformer expert but I represented them as best I could, the load current is approx. 3 Amps

The plot shown is the secondary current.

Approx. 65.5VA

Lawrence.

[kalee20]

That looks highly credible! (though 804mΩ for secondary resistance seems a bit high for a 40VA transformer... if you've measured one though, I'll take that back!)

Can you also show the capacitor voltage?

[Radio Wrangler]

The tolerance bounds on 115v supplies are not proportional to the tolerance bounds on 230v supplies.

If the design of both transformers has been done thoroughly, there may well be a slight difference in turns ratio to accommodate this for when the split primary version is in one mode or the other.

Such a difference would leave a small emf driving current around the loop created by the paralleled primaries of the two transformers.

Transformers aren't simply voltage ratio = turns ratio, it is normal to have correction factors to make up for resistive losses and load current. On top of this are factors relating to output specifications and what conditions they apply over.

David

[ms660]

Originally Posted by kalee20

That looks highly credible! (though 804mΩ for secondary resistance seems a bit high for a 40VA transformer... if you've measured one though, I'll take that back!)

Can you also show the capacitor voltage?

The mean voltage across the capacitor is 15.95 VDC as shown in the table in the screen shot in this and my previous post, enclosed is a screen shot of the voltage waveform across the capacitor, the ripple factor is approx. 6.5% according to my calcs.

Rightly or wrongly here's what I did, I inputted the OP's transformer off load voltage (17.5 VAC) its nominal voltage (15 VAC) an IAC of 0.666 Amps (10VA/15 VAC) and a regulation figure of approx. 14.28%, that spewed out a DCR figure of 3.216 Ohms, I divided that by 4 to get 0.804 Ohms, I multiplied the current of 0.666 Amps by 4 which gave 2.664 Amps.

Lawrence.

[kalee20]

OK - In the absence of anything better, it would be difficult to disagree with those figures - on the contrary, you could hardly do any better!

The ripple is a bit big. Upping the capacitance will reduce it, and hopefully demonstrate that the secondary current hardly changes.

The 65.5VA... is this calculated from the RMS value of the peaky current waveform and the nominal 15V secondary voltage? If so, it's certainly showing that the transformer throughput is a lot more than the rated 40VA that the OP has at his disposal, with the 3A DC load.

[ms660]

Originally Posted by kalee20

OK - In the absence of anything better, it would be difficult to disagree with those figures - on the contrary, you could hardly do any better!

The ripple is a bit big. Upping the capacitance will reduce it, and hopefully demonstrate that the secondary current hardly changes.

The 65.5VA... is this calculated from the RMS value of the peaky current waveform and the nominal 15V secondary voltage? If so, it's certainly showing that the transformer throughput is a lot more than the rated 40VA that the OP has at his disposal, with the 3A DC load.

Apologies for the delay in replying, been doing the hospital run most of the day, then cook tea etc.

The 65.5 VA I calculated was calculated from the PSUD2 results using the transformers RMS secondary voltage under load and the transformers RMS secondary current which is the diode(s) RMS current multiplied by the square root of two.

I also calculated the VA using Schades (just for the heck of it) for the transformers secondary current and using the results of PSUD2 for the unloaded and loaded secondary voltages and DC output voltage, I calculated the figure to be approx. 63 VA so not far out when compared to PSUD2.

The product of 2Pi*f*C*RL is quite low but the series resistance/load resistance percentage is quite high which means that the Shades curve is reasonably flat at those values.

Lawrence.

[kalee20]

Nothing amiss with your methods Lawrence - that, and everything, indicates the OP's transformers will be over-run!