Modifying a light dimmer output waveform to power a 240 volt transformer?

Thread Starter

FloWolF

Joined Nov 17, 2022
9
Hi folks, my first post here (waves!).

I'm knocking together a little power unit (~8 to 25 volts, 3-5 amps) for electro-etching, where the output needs to be switched from DC to AC during the process, and this bit of course is easy.

My problems start when I want to be able to control/adjust the unit output voltage for both AC and DC. With some trepidation I tried one of those low cost 240v/2kW 'motor controller' dimmers from ebay on the transformer input since I had a couple to hand, and it all seemed to be working OK, panel meter reading the voltage correctly etc. and I thought I'd gotten away with it.

In fact I was just about to hot air gun all my heat shrink and fasten up the enclosure before I realised I hadn't done a load test - made some saline soaked tissue as the etch simulation and pushed the etch wand down, and the panel meter light went out before any current draw was even displayed. Checking the transformer AC out showed me the transformer was not supporting the load with the voltage dropping from ~24 to about 1.3, and my fears about the dimmer output waveform being no good for driving the transformer were realised.

There's no way I can afford to buy a sine-to-sine wave dimmer, and who would for such a little project, and no way I'm going to start trying to build a dual IGBT unit myself or anything (I used to be a little better at this sort of thing, but my brain is going to mush these days), so I'm wondering if a simple solution like a capacitor alone across the output of the dimmer/input of the transformer might not do enough for the waveform to render it more palatable to the transformer? If so what type and value should I be in the ballpark with?

Help!

Thanks in advance, and Kind Regards,

Shaun/FloWolF
 

Ian0

Joined Aug 7, 2020
6,706
Try a dimmer for theatre lights. They are often used with low-voltage halogen lamps with transformers built into the luminaire.
 

Thread Starter

FloWolF

Joined Nov 17, 2022
9
Try a dimmer for theatre lights. They are often used with low-voltage halogen lamps with transformers built into the luminaire.
Thanks for the quick input, much appreciated.

My search for stage/theatre lighting dimmers didn't turn up anything I could use though, only large expensive units.

Cheers!

Shaun/FloWolF
 

LowQCab

Joined Nov 6, 2012
2,644
You may have made a hasty conclusion.

The Transformer may get hotter than normal,
( when "dimmed" by an SCR ),
but it should still work short term.

The Output-Voltage will have a lot of "trash", or "noise", added to it by the Switching action of the SCR,
this may fool your Meter into thinking that the Voltage is higher than it really is.

You need to make sure that You measure the Voltage/Current while under the full expected Load.
.
.
.
 

Ian0

Joined Aug 7, 2020
6,706
You may have made a hasty conclusion.

The Transformer may get hotter than normal,
( when "dimmed" by an SCR ),
but it should still work short term.

The Output-Voltage will have a lot of "trash", or "noise", added to it by the Switching action of the SCR,
this may fool your Meter into thinking that the Voltage is higher than it really is.

You need to make sure that You measure the Voltage/Current while under the full expected Load.
.
.
.
I can't think that the HF content of the step-function that is convolved with the sinewave will make much difference to the core heating. Asymmetry in the waveform may be more of a problem.
The copper loss will be reduced as the transformer is not being used at full power.
There used to be a disco lamp called a pinspot (which was a 6V 30W #4515 lamp housed with a transformer). They became a commodity product and there was a race to the bottom in price with transformer quality suffering.
However, reducing the transformer size by running them closer to the saturation limit made the transformers run hotter and saturate more on switch-on. Flashing the pinspots rapidly killed them, due to the saturation. Those on dimmers had no problems.
 

Thread Starter

FloWolF

Joined Nov 17, 2022
9
You may have made a hasty conclusion.

The Transformer may get hotter than normal,
( when "dimmed" by an SCR ),
but it should still work short term.

The Output-Voltage will have a lot of "trash", or "noise", added to it by the Switching action of the SCR,
this may fool your Meter into thinking that the Voltage is higher than it really is.

You need to make sure that You measure the Voltage/Current while under the full expected Load.
.
.
.
I wish but not in this case - The off-load voltages were checked with 2 different meters, and both agreed with the expected voltage of around 28VDC with a capacitor smoothing the output.

I tested under full expected load, and the voltage crashed to below 2 volts, and this reading was also evident right at the transformer output showing it was the transformer itself and not an issue with the bridge.

If there's a square wave component then the transformer winding is going to be seeing periods of DC, which it cannot work with, is this not right?

So back to my question, does anyone think a capacitor across the dimmer output would smooth the waveform enough to remove the DC elements and help the transformer, and if so what type and value would be in the ballpark?

Thanks again!

Shaun/FloWolF
 

Thread Starter

FloWolF

Joined Nov 17, 2022
9
I can't think that the HF content of the step-function that is convolved with the sinewave will make much difference to the core heating. Asymmetry in the waveform may be more of a problem.
The copper loss will be reduced as the transformer is not being used at full power.
There used to be a disco lamp called a pinspot (which was a 6V 30W #4515 lamp housed with a transformer). They became a commodity product and there was a race to the bottom in price with transformer quality suffering.
However, reducing the transformer size by running them closer to the saturation limit made the transformers run hotter and saturate more on switch-on. Flashing the pinspots rapidly killed them, due to the saturation. Those on dimmers had no problems.
The mains transformer will be optimised for the region of frequencies around 50Hz, so higher frequency signal components may greatly reduce efficiency perhaps. I don't think it's so much the HF content as the DC content?

So, what about that idea of a capacitor across the transformer primary?..

Cheers!
 

Ian0

Joined Aug 7, 2020
6,706
The mains transformer will be optimised for the region of frequencies around 50Hz, so higher frequency signal components may greatly reduce efficiency perhaps. I don't think it's so much the HF content as the DC content?

So, what about that idea of a capacitor across the transformer primary?..

Cheers!
It does nothing for the DC content. In conjunction with a series inductor it would remove some hf, but I don‘t suspect the amount of hf is a problem..
 

Thread Starter

FloWolF

Joined Nov 17, 2022
9
It does nothing for the DC content. In conjunction with a series inductor it would remove some hf, but I don‘t suspect the amount of hf is a problem..
Should the capacitor not charge in one direction, then discharge and recharge in the other, removing the DC and converting it into some shape of slope instead? I'm not looking to convert it to a sine wave, just modifying the waveform so the transformer is happy with it. Cheers!

Shaun/FloWolF
 

Ian0

Joined Aug 7, 2020
6,706
Should the capacitor not charge in one direction, then discharge and recharge in the other, removing the DC and converting it into some shape of slope instead?

Shaun/FloWolF
No. Does a smoothing capacitor remove the DC content of your power supply?
 

Thread Starter

FloWolF

Joined Nov 17, 2022
9
No. Does a smoothing capacitor remove the DC content of your power supply?
No, but it does charge and discharge to reduce the peaks and fill in the troughs, and simple RC networks can be used to approximate a sine wave from a square wave signal, and this is where I was coming from.

Understand I am not trying to be argumentative, but in order for me to shift old faulty knowledge and reasoning out I need to grasp and understand the new information, and where I got it wrong or it just won't sit. This isn't on purpose, it's just the way my somewhat autistic mind has always been, sometimes to my benefit, sometimes to my detriment.

Thanks for your understanding.

Shaun/FloWolF
 
Top