Transformer with dual primaries wired to produce positive & negative rails?

Thread Starter

Standisher

Joined Jan 16, 2015
156
At this stage of my learning I'm still a bit vague about wiring of transformers and what is possible. I have successfully removed windings to lower output voltage and have also successfully wired twin secondaries in series and parallel. I still have lying around a toroidal with twin 18VAC secondaries and wondered whether this can be used, like a centre tapped secondary to produce positive and negative rails (i.e 18V -0V -18V) ?

I've done lots of internet research and most articles suggest that positive and negative rails can only be created using centre-tapped transformers. I have however found two brief articles that suggest I can do it with twin secondaries. The transformer I have is this one of these (the MCTA160/18 variant). I am attaching an image showing, on the left the normal secondary arrangement and, on the right the configuration which, it is suggested, will produce an 18V -0V -18V output. I would be grateful if you knowledgeable folk could tell me whether this is possible and, if so, are there are any traps I need to be aware of.

If it is possible I would also need to try and get my head around how the rectification would be done.
transformer config.jpg

Regards
 

bertus

Joined Apr 5, 2008
22,270
Hello,

If the coils are connected correctly, the two voltages will be out of phase on the + and - 18 as you mention them.
In schematics you will see a dot at the coils to indicate the start of the coil:

ILP_connections.jpg

Bertus
 
This http://www.hammondmfg.com/pdf/5c007.pdf note can either make things more complicated or clear things up a bit.

A 18 VAC secondary will produce a DC voltage ABOUT the SQRT(2) higher than the AC voltage. The diode drop(s) have to be considered sometimes. Note from the hammod note that the DC current available from the rectified AC isn;t the same as the RMS AC current of the transformer.

There really isn't a voltage of -18 VAC. There is a direction. A Center tapped configuration like yours is usually written as 18-0-18.

When you connect the 18 VAC windings in series, you will either get about 0 Volts or a doubling. This allows you to determine the "phase".
In the proper configuration, you want about 36 VAC from end to end, not zero.

When you parallel windings, you need to pay attention to the phase and they need to be indentical windings, at least in voltage. You can use the trick above to determine the phase and then connect them phase to phase, otherwise they will smoke.

Doing primary identification is a bit harder, but not much. If you have two 0-120 primary windings and you live where there is 240, you have an issue. Just use any source voltage <=120 V. Primaries can be tapped; e.g. 0-120-240 etc or two 0-120 for example. Parallel windings double the current. The bi-filar technique where two wires are wound together makes the dual 120 V windings less costly in time to wind.
 
Last edited:

Thread Starter

Standisher

Joined Jan 16, 2015
156
Thank you @bertus and @shortbus for your helpful replies.

Just want to clarify:
Hello,

If the coils are connected correctly, the two voltages will be out of phase on the + and - 18 as you mention them.
In schematics you will see a dot at the coils to indicate the start of the coil:

Bertus
Are you saying that the diagram on the bottom right of your attachment is the correct method of connecting (i.e deliberately out of phase)? I've also attached an image lifted from my toroidal datasheet (which shows the 'dot' markings) annotated with my understanding of the correct connection for my application. Would be grateful if you could let me know whether I seem to have it right.

Here is a good link that explains it. There are also two ways to do what your asking, half wave and full wave rectification from AC to DC. http://williamson-labs.com/powersupply.htm Hope this is what your asking.
That is also helpful @shortbus as it seems to confirm, in the diagram entitled 'Dual Voltage Supply' my initial guess on how I should rectify (see image)

The transformer is rated at 160VA so I'm also guessing that the max current I can expect in this configuration is half that?toroida2.png twin secondaries2.png

Once again, many thanks for your help guys; much appreciated.
 

Thread Starter

Standisher

Joined Jan 16, 2015
156
Hello,

When rectified and smoothed, the output voltage will be about 18 * (sqrt 2) = 25 Volts.

Bertus
Hi,
Yep, thats what I was expecting. I'm trying to set up a dual, variable supply (upto +18 & - 18V rails would be excellent) for audio circuits. By the way could you give your thoughts on the clarification I was seeking in post #5 ? Appreciated.
Erik
 

bertus

Joined Apr 5, 2008
22,270
Hello,

Did you see KeepItSimpleStupid's remark

When you connect the 18 VAC windings in series, you will either get about 0 Volts or a doubling. This allows you to determine the "phase".
In the proper configuration, you want about 36 VAC from end to end, not zero.
That will give you the correct connections.

Bertus
 

crutschow

Joined Mar 14, 2008
34,281
Correct way yes, your transformer is rated at 2X 80Va @18V, thats about 4amps each output,

total is 36V @4amps max ..
That's for a sinewave RMS output current.
Due to the high peak (and RMS) transformer current of a rectifier-capacitor output, you should derate to about 55% of that or 1.1Adc maximum rectified dc current for each output.
This is shown here:
 

Thread Starter

Standisher

Joined Jan 16, 2015
156
That's for a sinewave RMS output current.
Due to the high peak (and RMS) transformer current of a rectifier-capacitor output, you should derate to about 55% of that or 1.1Adc maximum rectified dc current for each output.
OK, I'm just a bit confused here (nothing unusual with me :)) . I understand that I need to derate to about 55% but I don't get the bit about "...or 1.1A DC maximum rectified dc current for each output.". The transformer is 160VA or I'm thinking 80VA per secondary. I would have thought derating to 55% would give me about 40VA per output (or at 18V - circa 2.2A). I'm obviously being a bit slow here because studying the image you included has not helped me get my head around the low current I can expect (although it is probably OK for my needs). Is there any other way you can explain the math to help me understand? I do appreciate your input @crutschow
 
Last edited:

crutschow

Joined Mar 14, 2008
34,281
Sorry for the confusion. Your are correct. :oops:
I was somehow thinking each output was 2Arms.
Each output is rated at 80VA/18 = 4.44Arms
Derating that by a factor of 1/1.8 gives a maximum output for each voltage of 2.5Adc.
 
Last edited:

#12

Joined Nov 30, 2010
18,224
As a crude way of understanding the amperage de-rating effect, a capacitor charges to the peak voltage of the AC wave. The current delivered to the load, times that peak voltage, is still related to the watt budget. Here's chart that is an excellent piece of reference material for the various ways to use a transformer for basic DC power. It includes considerations more complicated than the obvious parts we are typing about.
 

Attachments

Thread Starter

Standisher

Joined Jan 16, 2015
156
Here's chart that is an excellent piece of reference material for the various ways to use a transformer for basic DC power. It includes considerations more complicated than the obvious parts we are typing about.
That is a good reference note @#12 , I was studying it earlier when @KeepItSimpleStupid posted it. Cheers.

Sorry for the confusion. Your are correct. :oops:
I was somehow thinking each output was 2Arms.
Each output is rated at 80VA/18 = 4.44Arms
Derating that by a factor of 1/1.8 gives a maximum output for each voltage of 2.5Adc.
Many thanks @crutschow .... appreciate that.
 
Intuitively, the rectifier flips the negative half cycle and makes the frequency 120 Hz. The capacitor has to "fill in the peaks". So based on the capacitor size and the load, the capacitor discharges a bit, so you end up with a DC with a low height squarish wave on tip of it where it;s initially charged to the peak voltage. That cap can't really keep up.
 

Thread Starter

Standisher

Joined Jan 16, 2015
156
Understand that (I think) @KeepItSimpleStupid ..... as we are 50Hz mains frequency in the UK, am I right to assume the rectified frequency becomes 100Hz? Not that I suspect that changes anything in the point you are explaining; just like to make sure I'm fully understanding everything (even as a kid I was told I asked too many questions ;) )
 

Thread Starter

Standisher

Joined Jan 16, 2015
156
Actually, just noticed I got the heading wrong in this thread Doh! I meant to say 'Transformer with dual Secondaries...' not 'Primaries' (it does actually have dual primaries but that was not an issue that concerned me). Sorry if that title was/is misleading to anyone. :oops:
 

#12

Joined Nov 30, 2010
18,224
Yes, 50 full cycles per second produces 100 peaks per second.
And don't worry about the title. Transformers work both ways (if you're careful not to overload them) so people who are familiar think in either direction easily enough.
 
Top