Restricting current flow?

[minimutly]

Care to expand on that? I've tried to keep it in simple terms rather than going into things like gate pulses as that level of details isn't needed at this point.

There's a document above which goes into more detail if needed, but the broad answer is that, yes you can restrict current with the proper components.

No, take it as me not reading down properly...

 

[sako243]

100 times by my reckoning.

Yeah we make them so I know a little bit about them….

View attachment 488537

You learn something new everyday

 

[daleyd]

You learn something new everyday

It's not really important as both terms are used pretty much interchangeably nowadays, I only know about it because it's on the bottom of a training slide we use in work so just thought it was a bit of useless trivia to throw into the mix!

 

[hotponyshoes]

Thanks all, this is useful information.

But, gives me more questions!

To keep things simple I'm going to start with one control/distribution box.
2× 16a outlets. One dedicated to the water heater. One for everything else.
Both sockets to supply 16a but the heater socket controlled by the total load on the system so the total never exceeds 20a.
Essentially, a 'smarter' version of the one I have just made.

(SCR means thyristor and is easier to type). Its the same as a large light dimmer, works exactly the same way.
SCR's chop up the AC wave form and reduce the time that an item gets power.
It can cause harsh interference and also lowers the power factor, drawing current when the voltage is low.

Yes the SCR needs to be bigger than the load current, however you could possibly parallel up a second SCR and drive it from the modules output to increase the rating.

This sounds a bit like pwm (which I have messed with on dc systems)
If I feed a 20a heater with pwm at 50% it gets 10a. But, is that actually 10a of current or is it 20a pluses for half of the time?
If it's 20a pulses won't it still trip the generator on the first pulse?

What sort of interference does it cause?
Anything that's going to interface with other devices on the same system?
I'm not worried about picking up noises on the radio! (Although the card machines run over wifi)

I'm not sure if power factor is an issue as the hearer will only be using 'excess' power.
Ie, if I have 4000w 'spare' it would be better to use 3200w of that than nothing or 500w in a smaller heater.

An SCR stands for silicon control rectifier (not a trade name, just an abbreviation). It is to all intents and purposes a switchable diode (diodes are silicon devices used for rectifying current hence the name). Diodes allow current to flow in one direction only (within the device limits). Normally you can't turn them on or off, they're just always on.

An SCR is turned on (allows current to flow) when a current is passed into the gate terminal. You only need a pulse, they remain conducting as long as the current flowing through the main anode/cathode junction remains above the holding current. I.e. The only way you can turn them off is to remove the load.

They don't work well for AC because they only allow current in one direction so will inherently rectify the current flowing through them. A TRIAC is another form of thyristor that is effectively two SCRs back to back. This allows current to flow in both directions through the main body of the device. These need a constant gate current to stay on because at some point during the AC cycle the current will drop to zero briefly (50 times a second).

To limit "power" through the device normally you have some electronics synchronised to the AC waveform with a zero crossing detector. You then turn on the device a controlled period after the zero crossing (as it won't turn off until the current flowing through it is zero). The relationship isn't quite linear as @daleyd says because the waveform is sinusoidal but it's close enough most of the time.

Does an scr on ac power turn off then? If I pulse the gate will it turn off 1/50 of a second later?
Are 16a triacs easily available? A din rail mount would be ideal..

Regarding control,
I found this thing

That's a cheap chinese thing but I'd imagine a branded equivalent exists.

That can be set to monitor 0-20a (perfect) and gives a 0-10v dc output.
Can I power a gate with dc?

That 0-10v will be the wrong way around for me. 10-0v versions may exist or I'm sure I've seen (may even have) a reverser that takes 0-10v and turns it around ie. 9v In = 1v out.

Could control be as simple as running the incoming supply through that and driving the gate from it?

 

[Robotstar5]

An SCR is a trade name for a specific type of thyristor, but the terms are broadly interchangeable. Typically to control current in both directions you might have 2 SCR back to back, or might use a TRIAC - both of which belong to the thyristor family of devices. The company I work for makes thyristors but we sell than as those, not SCR - though they may contain an SCR inside, if that makes sense.

Well you learn something every day, I've spent my whole working life thinking it was a Silicon Control Rectifier, we used a lot of them back to back in spot welders where they did clever things chopping the power part way through the cycle to control power to the welding transformer. They were certainly an improvement over the mercury arc rectifiers used in the previous generation welders, when I started in the 70's we still had some in the stores!.

 

[daleyd]

You would need something in between to do some sort of control, to monitor the feedback and provide the correct output to the Thyristor(SCR if you prefer)

In phase angle control, the thyristor chops up the waveform so controlling the voltage (And by default the current) to the load.

Here's another PDF showing the variations on how thyristors can be used to control - you can see that Burst mode is more like PWM in that it provides full current for varying periods, to give an "average" power out.

Only phase angle gives the option of current control. But comes with disadvantages of being electrically "noisy". Lots of industrial applications prefer Burst mode as it is "cleaner".

I went to an oil platform in Holland a few years ago to modify a system to control on phase angle - it was a big 1000A gas dehydrator, and it was initially fitted to run in burst mode, as they were concerned about harmonics being fed back to the onboard generators - but when it conducted you could hear the generator slowing down and the lights dimming as the load suddenly jumped!

Changed to phase angle control and it was much smoother, as it was able to control in a much more analogue fashion, rather than being on or off.

 

[daleyd]

Well you learn something every day, I've spent my whole working life thinking it was a Silicon Control Rectifier, we used a lot of them back to back in spot welders where they did clever things chopping the power part way through the cycle to control power to the welding transformer. They were certainly an improvement over the mercury arc rectifiers used in the previous generation welders, when I started in the 70's we still had some in the stores!.

I don't think it's one or the other - from my understanding we tend to refer to them as thyristors, possibly because of the historic association as another companies trade name, or maybe the fact that it may be another type of device inside (TRIAC etc) but everyone seems to use the terms interchangeably.

From a definition point of view all SCR fall under the thyristor umbrella - but not all thyristors are SCR's (If that makes sense!).