Using Two Identical transformer to form Centre Tap Configuration

[basit701]

Hello Guys!

My requirement for project is 24_0_24 volt and 2amp each tap. Unfortunately 24_0_24 volt center tap in my area is only available with current rating of 800mA. To meet the 2amp requirement I bought TWO 24v 2 amp transformers and connected their outputs in series to form 24_0_24. The two transformers are identical i.e. same manufacturer, same rating etc.

The two transformer configuration is working perfectly, but when I discussed about this method with my friends. Their reply was not in favour of this, most of them said that if one of transformer gets short circuited it will short the other transformer and both will not work. A couple of them said that it will damage the components like ICs etc.

What are they saying is correct or not?

[Richard Crowley]

Your solution is perfectly fine. Your friends are misinformed.

[Kleinstein]

The solution is fine, even if the transformers are not identical. The downside is, that 2 smaller transformers usually are larger, more expensive and have slightly lower efficiency.

[basit701]

So this configuration will work fine as centre tap will?

[SeanB]

Yes, it will work. I have taken apart some OKI printers, and there they used 2 separate 120VAC transformers, with the mains voltage switching being to connect the primaries in series for 230VAC and in parallel for 120VAC use. Secondaries were used like you did to make a centre tapped winding giving 24VDC for the hammer drives, and another 2 secondaries to give 8VDC for the 5v rail and 12V rails for the RS232 interface.

[retrolefty]

So this configuration will work fine as centre tap will?

 Yes it will. However be aware you have to make sure you wire the two secondary windings so they 'add' and not 'subtract' causing zero voltage output.

[Gyro]

Small point... It does depend a little on what you are doing with the outputs. If you are planning to put a bridge rectifier across the two 24V outputs to create a center tapped DC output (as I suspect you are) then you'll be fine as long as the load on both DC rails is equal.

If you were planning to use the transformers in a traditional 'full-wave' configuration - 2 diodes, to generate a single rail then you would get DC saturation of the transformers.

As I say, probably not the case here, but something to be aware of if blanket replacing a single transformer with two.

[IanB]

If you were planning to use the transformers in a traditional 'full-wave' configuration - 2 diodes, to generate a single rail then you would get DC saturation of the transformers.

As I say, probably not the case here, but something to be aware of if blanket replacing a single transformer with two.

Why would two transformers in series behave differently than a single centre tapped transformer in this configuration? Is it because of unbalanced current in the primaries?

[CatalinaWOW]

For what it is worth, your friends are at least partly right.  If the primary of one transformer shorts it will kill both outputs.  So what.  If you had a single center tapped transformer and shorted the primary it wouldn't work either. 

[Gyro]

Yes, in a single transformer (ie. single magnetic core) then imbalances between the secondary windings cancels. Not necessarily the case if you have 2 separate cores.

Somebody please tell me I'm talking rubbish here 

[Richard Crowley]

Yes, in a single transformer (ie. single magnetic core) then imbalances between the secondary windings cancels. Not necessarily the case if you have 2 separate cores.

Somebody please tell me I'm talking rubbish here 

How is that any different than using a single transformer with a single half-wave rectifier?  Its been done a million times with no apparent ill effects.

And using dual transformers like @basit701 proposes is done all the time as well.  I don't see what all the fuss is about?

[Gyro]

I am probably wrong  - I just can't think whether an unbalanced current draw after the (presumed) bridge rectifier relative to the center tap would cause any DC saturation in the transformers.

Using a separate bridge on each transformer and combining the center tap at the DC side avoid any potential issue anyway.

[Gyro]

How is that any different than using a single transformer with a single half-wave rectifier?  Its been done a million times with no apparent ill effects.

Yes it has been done a million times, but half wave rectifying the output of a transformer seriously de-rates its output current vs full wave rectification.

[basit701]

I am using this rectifier circuit.

[Gyro]

This is the approach I was suggesting (ignore extra components, I just pulled it off Googe images).

It may be overkill, but It addresses the possible issue that I raised.

[basit701]

This is the approach I was suggesting (ignore extra components, I just pulled it off Googe images).

It may be overkill, but It addresses the possible issue that I raised.

The circuit you posted contains independent bridge rectifiers for each tap. What the significance of using this over a single bridge rectifier for both taps as in the circuit which i have posted.

[eneuro]

What the significance of using this over a single bridge rectifier for both taps as in the circuit which i have posted.

For example in your single bridge design if one of the capacitors will change its capacity due to many factors than you can end up with different +V and -V  since they are charged in series-additionaly one of them can be overcharged to higher voltage than nominal if this capacity change will be higher

So, in simple mosfet driver design like this where I have two caps in series and ~24V accross them, I have additional resistors to balance capacitors voltage and keep its voltage at the same +/-V 
But you have a few amps outputs there, so such simply balancing is not an option, so you'll probably have to pray and keep those caps at good condition, so that is one of the reasons I better like solution with two bridges and separate floating PSUs to form +/-V 

[Gyro]

Sometimes, especially in a one-off application it is simpler to do something that you know will absolutely work than spend too much time wondering about a possible issue (like the possible DC imbalance with unequal load that I was talking about). I may well be completely wrong but I would need to analyse it in detail to be completely comfortable. As I say, I know for sure the dual bridge circuit will work as expected.

That said, feel free to make me slightly uncomfortable. 

[dom0]

You are aware that the dual-bridge-rectifier configuration is identical to the standard center tap configuration, except you need double the diodes and have doubled the rectifier losses?

@eneuro: this is false. The maximum potential imbalance across the filter caps is determined by the transformer, not their capacitance. What you say would be the case if you did not connect the center tap.

[Zero999]

The dual bridge rectifier circuit offers no benefits for a single, centre tapped transformer but has the benefit of preventing core saturation if two transformers are used and the DC currents are significantly unbalanced.

[eneuro]

As I say, I know for sure the dual bridge circuit will work as expected.

Yep, you have two floating PSUs.

Voltage imbalance I was talking above-probably this is not what we want-made quite bad scenario where one cap capacity changes +20% while other goes down -20%  and bingo +20V vs -30V


Fundamentals friday 

[eneuro]

The maximum potential imbalance across the filter caps is determined by the transformer, not their capacitance. What you say would be the case if you did not connect the center tap.

Haven't got more time today to put into simulator whole circuit, so you can be the first to proove you are right 

[dom0]

No simulator needed for that. 

[Gyro]

@dom0

I've gone from a nagging doubt to sketching this out on the back of an envelope....

For the single bridge rectifier case, take an extreme case of full load on the +ve rail and zero load on the -ve rail. Now look at the current waveforms (and therefore the core magnetisations) on each of the transformers. Now compare with the secondaries (overall magnetization) on a single core (ie single transformer). I think they are different.  What do you think?

[dom0]

I see your point there, for large imbalances that becomes an issue (or a really large, fuse-blowing one with toroidal transformers).