SCR Controllers 5

[Tlas3]

I got a question about SCR controllers and I was wondering if anyone could help me out.

I'm trying to control a large resistance heater to high accuracy using an SCR controller. Issues have been raise whether to use the phase angle fired or zero voltage cross.

I'm more inclined to use the phase angle fired because i believe it can reach higher accuracies since zero cross is bound by each period of the 50 sine waves (50hz). However i'm also aware that phase angle fired is predominantly reserved for high response heaters such as wires due to the high Di/Dt. I've also read that phase angle fired tends to have high EMI (electro magnetic interference) due to the fact that it creates harmonics.

I'm very open to those, opinions and suggestions for using a Phase angle fired SCR for a high load (over a two dozen KW) resistance heater.

 

[Skogsgurra]

Zero cross (sine wave packets) does not produce harmonics and HF EMI. But it does cause flicker, which can be a problem if you have a week mains.

If the heaters heat water, or some other large mass with high heat capacity, the zero crossing can give you good accuracy. But if you are producing heat in radiation form or heating thin/small/light things, then phase control is a must. The harmonics issue is not at all that bad. It is usually exaggerated and filters can reduce the problem if it gets real.

I would go for phase control.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Hello Tlas3.

As skogs noted, normally phase angle is pretty much reserved for RADIANT heating. I do much work in the semiconductor equipment realm and they use some very large heaters in many applications that require insane temperature control. The only place we bother with phase angle(PA) control is with radiant heaters. Essentially lamps.

I somewhat disagree with skogs in that PA can and does cause far more EMI then zero crossing(ZC) - sometimes in mind numbing quantities too.

If you are not using radiant heating I cannot imagine you even being able to detect the difference. Except in your wallet. The key is to keep the cycle time reasonable.

There is an abyss between ZC and PA you cannot reasonably go much shorter than a second for ZC because the control become quantized due to the 1/2 cycle limit imposed on the power.

There are random SSRs and then there are PA controllers don't confuse them.

PA controllers are probably 3X the cost of ZC.

What are you heating?
What precisely are your heating elements?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jraef]

Another important aspect is in your system design. If you are using a transformer downstream of the SCR controller, you cannot use Zero-Cross controllers, only phase angle. Also, if your heating element has a high hot-to-cold resistance ratio, such as quartz lamps do, it is best to ramp the power into it to avoid damage, something that can be done best with phase angle control. There are a few people who make hybrid systems, i.e. phase angle to ramp up then change to zero-cross for control to avoid interference. This is what is used in semiconductor manufacturing machines where high precision is an absolute must, so the accuracy is not inherently bad for zero-cross, it just takes some attention to detail in manufacturing the trigger system.

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[Skogsgurra]

Keith, I said Harmonics - not EMI (HF). We do agree.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

You're correct. Leave it to you to actually carefully read each word.. I think you missed your calling skogs.

Speaking of flickering. In debugging a controller I built I needed a load. I grabbed a quartz lamp I'd ripped from a laserjet fuser (1kW) and hooked it to my SSR (ZC). I then proceeded to dink with the cycle times and set points and other parameters. At one point I was shocked to find the bulb, which was rather obvious, hanging from wires from my desk structure (cuz it was HOT!) shift from the expected stroboscopic flickering to long on and off periods of perhaps several seconds.

This was caused by the control signal beating with the turn on region of the ZC function. Since the ZC function limits the ON trigger period to about 20 degrees of the increasing line cycle.

I figured that there must be the occasional engineer out there driven nuts by a control system that was running thru these beating periods and not understanding why the system might just get into the setpoint region then suddenly have lousy swinging process temperatures and since normally you aren't using quartz heaters never 'see' the problem.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[mc5w]

A lot of electrical utilities have banned zero crossing a.k.a. burst cycle controllers because these produce modulation sidebands similar to what a radiotelegraph transmitter has in its output signal. If a burst cycle controller is running 1 cycle on 1 cycle off it produces strong modulation beat frequencies at 30, 90, 120, and 150 Hertz in the case of a 60 Hertz power system. These modulation frequencies create a lot of power quality problems such a flicker that are almost impossible to filter unless there is another burst cycle controller with an identical heater that operates in opposition with the first heater.

In acutality, if your heater is small, you could get away with a burst cycle controller without the electrical utility noticing. Depends on how much the neighbors complain about lights flickering and electronics malfunctioning.

Phase displacement controller power quality problems on the other hand can be tamed with a band pass filter, harmonic traps, and a rapid response capacitor bank to correct lagging vars.

Mike Cole

 

[Tlas3]

Thank you very much for all your comments. They have all been extremely informative. Many of your replies cleared up a lot of things for me. Thank you all

 

[itsmoked]

Slow reader? Oh no no! You only read the forum once per month on about the 18th!!



Kidding aside thanks for the come-back.


Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Tlas3]

Haha! I bide my time for the right time of the month but seeing as how I don't want to wait until next year i'll reply now =)

Nah, I've always intended to reply asap but slipped my mind. When I went through my email and saw the old scr controller threads I thought to myself it would be terribly bad manners of me not to thank you... despite being very late. So yeah i do feel bad it's been so long but truly, thank you everyone.

In the end, we purchased a phase angle fired SCR controller to control airflow using a high load resistance heater. I'm now responsible to test the minimum graduations of the controller. But since it's a very high voltage i'm hesitant as to where to begin. Perhaps there's a way to testing it using lower voltages input, control and output.

Why they have a Mech doing the work of an electrical or power engineer i haven't a clue...

'I wonder what happens when i connect these two wires toget-'

 

[Tlas3]

I also forgot to mention:

Have a Happy and safe Holiday!

 

[davidbeach]

Nah, I'm sure its not very high voltage (which isn't even defined) since you have to get to 1000V before it isn't low voltage; medium voltage is from 1kV to 69kV and then its high voltage until extra high and ultra high. So, I reasonably sure you are only working with low voltage.

 

[Skogsgurra]

The formal definition of High Voltage does not make Low Voltage less dangerous. I have never been able to sweet-talk a 400 V system into being less nasty to touch! ;-)

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[davidbeach]

skogs, I absolutely agree, low voltage is more then enough to do serious damage. It's just a personal crusade of mine (tilting against windmills perhaps) to stop the nasty habit of some folks to refer to anything over 5V, 12V, 24V or maybe 48V as high voltage.

 

[ScottyUK]

david,

Do you hang out with instrument techs a lot? Over here they're known as Tiffies and the majority jokingly (I think ) plead that anything over 24V DC is high voltage.

British and probably European officialdom no longer recongnises anything other than LV ([≤]1000V AC / [≤]1500V DC) and HV (>1000V AC, >1500V DC). The industry terms of MV and EHV are still heard but are no longer recognised by legislation.

----------------------------------

Sometimes I only open my mouth to swap feet...

 

[Skogsgurra]

The habit of talking about mains voltage as "High Voltage" is a good warning sign. People that do that should seek help from us Sparkies - or they may not be able to talk at all some day.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[jraef]

Since the original thread question is already satisfied, I'll wax nostalgic.

I once hired a woman (although this is not gender related IMHO) still in school at UC Berkeley for her EE degree, 3rd year. Her summer job was to just draft up control schematics for motor control panels that I had already designed. 99% of our panels used 24VDC controls. After a week of so she came to me asking "What are these circuits doing, because to me they don't make any sense. For instance I don't see what you could possibly do with these all series capacitors and varistors!" Being that this was all relay logic, I was puzzled by her question, then came to find out that she was interpreting N.O. contacts as capacitor symbols and N.C. contacts as varistors. I said "I thought you were an EE student?" She replied "Well, EE / Computer Science. We never work with this "High Voltage" stuff in class." To her, High Voltage was anything above TTL!

This was in 1983 and represented my first exposure to the fact that University EE programs were tuning out specialists who were no longer being exposed to the full gamut of EE disciplines.


JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[davidbeach]

Very few electrical engineers being produced these days; just lots of electronic/computer engineers who think that the electrons that make their creations work just magically appear when a cord is plugged into the wall. Electrical engineering seems to be a dying art.

 

[Skogsgurra]

Same thing here. "If it ain't 'puter it don't mean a thing"

It is OK for me - I make some money out of other peoples ignorance. But it is bad for society.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...