Using a transformer in reverse

Thread Starter

JulesP

Joined Dec 7, 2018
383
I would like to convert 230 mains voltage to 400 and then rectify it to produce 400V DC.
Transformers for that are not common but I have found one that can take 415V in and produce 240V out. Could I simply use this in reverse and then feed that to a standard full wave rectifier?

Thanks
 

Attachments

AlbertHall

Joined Jun 4, 2014
12,345
Yes, but...
415V rms will produce about 580V if you rectify it and add a smoothing capacitor.
Will you be smoothing the DC output of the bridge?
 

drc_567

Joined Dec 29, 2008
1,156
Due to the RMS meter measurement of the 230 mains voltage, in order to obtain 400 volts DC, you need about 283 RMS volts from the transformer secondary, as opposed to the 400 number. After rectifying and filtering, the 283 volts will be approximately 400 volts DC. Usually, there is not any problem with reversing the primary and secondary of a power transformer, as long as the transformer power rating, the product of RMS volts*amps, is not exceeded.
 

crutschow

Joined Mar 14, 2008
34,281
Yes, you can use the transformer in reverse, but note that the ratio is usually adjusted up by 5-10% to allow for the voltage loss under rated load due to winding resistance.
When you use it in reverse, that will result in a lower output voltage than expected.
 

TeeKay6

Joined Apr 20, 2019
573
I would like to convert 230 mains voltage to 400 and then rectify it to produce 400V DC.
Transformers for that are not common but I have found one that can take 415V in and produce 240V out. Could I simply use this in reverse and then feed that to a standard full wave rectifier?

Thanks
You can also simply use a transformer to create an isolated low voltage, e.g. 24VAC, and put that in series with your 220VAC source. Note that phasing matters; if wrong, the 24VAC will be subtracted, not added.
 
Hammond manufacturing has a nice pdf about rectifiers. When using the full wave bridge. the DC current available is MUCH less than the AC rating of the transformer.

As was mentioned, the 0 to peak or peak to peak value of the AC is what you get for DC.
 

Reloadron

Joined Jan 15, 2015
7,501
The same nice people at Hammond Manufacturing who published that Design Guide For Rectifier Use also still manufacture and market Power - High Voltage For General Amp Use Transformers. Consider your power requirements and maybe get a transformer designed around your intended application. Hammond is merely a single example, manufacturers like Stancore also come to mind. Anyway if you use a transformer in reverse make sure it will meet your power requirements.

Ron
 
How about a link to the Hammond webpage offering that "nice pdf"? Thanks!
I was using my phone and my phone is pre-historic. That's the reason you didn't get a link. Even at home, I have to install add-on to remove the Google wrapper.

Identical transformers can be paralled, but Nigel ran into one that was marked wrong. If you connect their secondaries in series first you get zero or double. The secondaries add or subtract. Once you know the phasing, you can place windings in parallel.
 

TeeKay6

Joined Apr 20, 2019
573
I was using my phone and my phone is pre-historic. That's the reason you didn't get a link. Even at home, I have to install add-on to remove the Google wrapper.

Identical transformers can be paralled, but Nigel ran into one that was marked wrong. If you connect their secondaries in series first you get zero or double. The secondaries add or subtract. Once you know the phasing, you can place windings in parallel.
@KeepItSimpleStupid
No problem. @MrChips was immediately online with the link.
 

MisterBill2

Joined Jan 23, 2018
18,167
I would like to convert 230 mains voltage to 400 and then rectify it to produce 400V DC.
Transformers for that are not common but I have found one that can take 415V in and produce 240V out. Could I simply use this in reverse and then feed that to a standard full wave rectifier?

Thanks
YES, you can use the transformer backwards and get the higher voltage, BUT the wattage will be the same, or possibly a bit less. And the rectified voltage when you add a filter capacitor will be quite a bit more than the rated 400 volts. So you will need a higher voltage capacitor. If you use a full wave bridge rectifier the DC voltage will be closer to 600 volts, which, by the way, is a dangerous voltage, so be very careful to avoid shocks. And if the transformer has an internal temperature fuse, it will not offer quite the same protection. BUT I have only seen such a fuse in transformers that I would call lower quality parts.
 

Thread Starter

JulesP

Joined Dec 7, 2018
383
Yes, but...
415V rms will produce about 580V if you rectify it and add a smoothing capacitor.
Will you be smoothing the DC output of the bridge?
Yes if you mean removing substantial ripple. I assume it’s always best to smooth unless one has clear reasons why one shouldn’t.
 

MrAl

Joined Jun 17, 2014
11,389
A couple of points about using a capacitor smoothed resistive load on a full wave bridge rectifier.

First, the average DC output voltage is not always 0.9 times the AC voltage. It depends on the ratio of the capacitor to the resistive load. For small capacitance and higher output current, the average DC voltage will be lower, and for large capacitance or lower current output the average DC voltage will be higher.
Consider for example a 100000uf cap and 1Megohm load. The average output DC voltage will be close to the peak voltage. For 0.001uf cap and 1 ohm load the average DC voltage will be closer to 0.9 times the AC rms voltage but of course will have lots of ripple.
That is with an idea transformer with 0 ohms internal resistance for both windings.
For real transformers with internal primary resistance Rp and secondary resistance Rs we can reflect the resistance of the primary to the secondary and that resistance adds to the secondary resistance. So you get two resistances in series on the output which can have a profound effect on the voltage drop.
For example, with 100 ohm primary and 1 ohm secondary and 10:1 ratio, the primary resistance reflected to the output is 1 ohm. That adds to the 1 ohm to get 2 ohms. So you have an ideal transformer with 2 ohms on the output secondary.
So with 1 amp output current the resistance will drop 2 volts.
The primary resistance reflected to the secondary in this case is 100*(1/10)^2 where 1/10 is the step down voltage turns ratio.
Of course the input voltage would have to be 100 volts and the load resistance 8 ohms to get a 2v drop on the output, but it's just an example. Now with either 100 ohms in the primary and 1 ohm in the secondary, or, 0 ohms in the primary and 2 ohms in the secondary (again 8 ohm load) the output voltage is 8 volts with either combination.
 
Last edited:

Thread Starter

JulesP

Joined Dec 7, 2018
383
It would probably help if I explained my application so as to clarify what I need. As with my posts earlier in the year, I’m researching the behaviour of short duration high voltage DC pulses.

In this case I'm trying to deliver 400V DC pulses delivered to a BLDC motor via an Arduino controller and interface, which uses an IRF840 mosfet for each of the three windings (Pic 1). The interface uses back-emf rotor feedback and the max power through the system is 0.8-1 kW.

It seems there are various options to arrive at the 400V DC before it is ‘handled’ by the controller.

One option is to take mains ac, up that a bit in a chunky power transformer to about 283V RMS (thanks drc_567) rectify and smooth that to arrive at the 400V DC.

Another option is to use a 12V car battery and a DC Chopper (e.g. https://s.click.aliexpress.com/e/1rHgPQC ) that produces a 400V 20kHz square wave output, that would presumably need to be rectified (and smoothed?) before feeding to the IRF840s in the interface controller.

A third is to feed a 12V battery into two inverters that each produce 220VAC, then rectify and smooth that (Pic 2).

Given the expense and bulk of a transformer for option 1, I’m inclined towards 2 or 3. I already have a car battery and a 500W inverter and getting another is relatively inexpensive and the rectification and smoothing (if required) is straightforward (using perhaps IN4148s and 500V 10uF caps?)

Jules
 

Attachments

Thread Starter

JulesP

Joined Dec 7, 2018
383
It would probably help if I explained my application so as to clarify what I need. As with my posts earlier in the year, I’m researching the behaviour of short duration high voltage DC pulses.

In this case I'm trying to deliver 400V DC pulses delivered to a BLDC motor via an Arduino controller and interface, which uses an IRF840 mosfet for each of the three windings (Pic 1). The interface uses back-emf rotor feedback and the max power through the system is 0.8-1 kW.

It seems there are various options to arrive at the 400V DC before it is ‘handled’ by the controller.

One option is to take mains ac, up that a bit in a chunky power transformer to about 283V RMS (thanks drc_567) rectify and smooth that to arrive at the 400V DC.

Another option is to use a 12V car battery and a DC Chopper (e.g. https://s.click.aliexpress.com/e/1rHgPQC ) that produces a 400V 20kHz square wave output, that would presumably need to be rectified (and smoothed?) before feeding to the IRF840s in the interface controller.

A third is to feed a 12V battery into two inverters that each produce 220VAC, then rectify and smooth that (Pic 2).

Given the expense and bulk of a transformer for option 1, I’m inclined towards 2 or 3. I already have a car battery and a 500W inverter and getting another is relatively inexpensive and the rectification and smoothing (if required) is straightforward (using perhaps IN4148s and 500V 10uF caps?)

Jules
Actually I’m going to have more than 440V with option 3 (x1.4) so maybe option 2 - smooth the dc chopper output
 

MisterBill2

Joined Jan 23, 2018
18,167
You will not have much except for smoke if you use a 1N4148 for the high voltage or high current rectifier. And I suppose that an inverter does make sense, but instead of two you could use a voltage doubler circuit. The voltage might still be a bit high, though.
Are you trying to drive the motor or just see the response to single pulses? How precise does the voltage control have to be for your application? Years back I ran a "fullwave" voltage doubler straight from the AC line and got 270 volts at no load and a lot less, about 95 volts, under a heavy load. It had very poor regulation, it appears.
 

Thread Starter

JulesP

Joined Dec 7, 2018
383
I'm planning to try running the BLDC with regular pulses from the Arduino and interface and returning to the question of how to get a smooth 400V DC, for various reasons I have decided to use the DC Chopper option so my query evolves into how to treat the 420V, 20kHz square wave output to derive a smooth 420V DC. Should I use 'Full Wave' rectification or some form of smoothing/filtering. I am presuming the square wave is not negative and positive but 0-positive only. (See pic)
 

Attachments

Top