Platen Heater Detection

On a Platen heater system what is a good way to have a heater diagnostic?

We have ammeters of each heater zone so we can see the current each zone is pulling but I need a way to know if it is pulling the right amout of current for the actual temperature at the time . some heaters zones are series and some paralle some a combination of both.

If you have a heater short out then that makes it real easy to find. But if a heater wired parrell becomes open you do not know and the other heats of that zone and other zones try to compensate.

What is a good way to tell if my heater circuit is working as it should based on the temp,rise time and amps it is pulling?

Or is it even possible?

Is each heater section powered by a variable current or a pulsed ON/OFF control? If the latter than, when turned ON each zone should pull a fixed amount of current. If one of a parallel group opens then the current will lessen. The PID controller may have to keep the zone on for a longer time in an attempt to make up but at any one moment the current will be less than it normally would be.

We have both types of systems Bernie.

Some are on/off ssr and some are variable current scr.

Would the heater pull the same amount of current at a cold platen state as they would if it were in 2-3 degrees of setpoint?

Setpiont on most are 220 C

If I can find a way to tell then I want to bring this info into wonderware for troubleshooting.

I guess on the on/off systems I could do logic that if the heat is on for that zone and the current being pulled is below limit then set an alert. Maybe I can use a limit instruction in the plc to do this feed the current ring analog value into it.

What do you think?

How could I approach the variable current system

If it's a PLC PID loop controlling the current then you would have access to the control output. Compare that, with appropriate math to adjust control output to get expected current (just measure with a properly running system to develop the conversion value), to the current.

The Plc Kid said:

On a Platen heater system what is a good way to have a heater diagnostic?

We have ammeters of each heater zone so we can see the current each zone is pulling but I need a way to know if it is pulling the right amout of current for the actual temperature at the time . some heaters zones are series and some paralle some a combination of both.

If you have a heater short out then that makes it real easy to find. But if a heater wired parrell becomes open you do not know and the other heats of that zone and other zones try to compensate.

What is a good way to tell if my heater circuit is working as it should based on the temp,rise time and amps it is pulling?

Or is it even possible?

Click to expand...

Well the old fashioned way was the operator complaining about it is not heat up fast enough (rise time of heated object). So you ran clampon readings to check if current is OK. IF low you had a burnt out heater. Then what you did was either take clamp on on each heater to find which one was burned out (or killed power and checked resistance).

Heater resistance will vary with temperature of the HEATER itself. This will vary somewhat proportional to what the heater is heating and type of heater ie
immersion would be more affected by liquid temp
than would be a heater inserted in a block of steel

IF you have only two or three heaters in parallel then +/- 10% current should be good enough.

I suppose with current transducers (either CT or Hall effect) with output compatible with PLC it would be possible to have PLC check for proper current. Ammeter would be cheaper I think - depends on how many heaters you have.

Dan Bentler

Ok this system now has a current transducer for each heater circuit.

I can now see actual amps on my versa view. I have all this programmed and working correctly.

Now I need 1 additional display percent of FLA. I have the FLA for each heater circuit but this system is variable voltage from PID so how would you do a percent of Fla?

On my on/off systems I just take actual amps and divide by known FLA (constant Value)

But with variable voltage doing it that way would not be correct right?

I suppose having teh PLC indicate actual amps or percent power would be kinda neat. Even if I had it I would still use a clampon to confirm heater operation - yes I know clampon would not be accurate on PWM so I would check on line side of controller. Power in = power out of controller so if line side amps and volts are OK you are OK.

You could also just put in Darsinol (spelling I know) movement meters - supposedly they operate on PWM just fine - have not tried.

Just because you have this neat gizmo does not mean you have to use it or much less trust it. I do not trust computer inventory until I check the shelf.
Operator does not care about the heaters - only about the temperature. I think for simplicity sake displaying only temp would be easiest.

Dan Bentler

I fully agree and would just like to display temp but in my defense it’s not my dog i.e. management wants it and very insistent on it

Temp is good enough for me amps was just a treat.

With these being variable voltage how far off would my amp reading be with the standard ct?

The Plc Kid said:

I fully agree and would just like to display temp but in my defense it’s not my dog i.e. management wants it and very insistent on it

Temp is good enough for me amps was just a treat.

With these being variable voltage how far off would my amp reading be with the standard ct?

Click to expand...

Amps are amps. A CT doesn't know about or care about the voltage. Just take your total measured amps and divide by the FLA rating.

Now, if you want percent _power_, that's a whole different animal. But, if it is management that wants a number, percent amps is a number.

If they actually want percent power, you will need to measure voltage too. At that point, I think I would just put in a power meter on the feed to the heater circuits and let it do all the heavy lifting. You can get one with comms and calculate a percent power to show on the screen.

Brian

The Plc Kid said:

I fully agree and would just like to display temp but in my defense it’s not my dog i.e. management wants it and very insistent on it

Temp is good enough for me amps was just a treat.

With these being variable voltage how far off would my amp reading be with the standard ct?

Click to expand...

Sometimes it is just like a two year old in the grocery store - it is easier to give in to managment.

The original idea I understood was to monitor power or current to ensure all heaters were operating ie no opens.

I am not sure if CTs will overheat at very low frequency.
On other hand they are used in VFDs and if you look at the current on a scope it is fairly sinusoidal so maybe CTs are just fine on PWM after all.

What gets me is
1. You are varying power by using PWM to reduce power as you approach setpoint and avoid overshoot.
2. At any time you are going to have X% of total available power being delivered based on how far you are from temperature setpoint ie SP - PV.
3. So are they looking for X% power or the percentage of X power to indicate all heaters are OK


And if you give them whatever the number they want - what are they gonna do with it?? Do they even know how much money they are spending to get some measurement they do not even understand much less know what to do with?? That just may be one of those mysteries understood only by management.

Dan Bentler

Just for educational purposes what if I had variable voltage/phase angle would that effect the accuracy of my ct with the ct I am using . This model says sinusoidal and non sinusoidal currents.

I think I am getting more lost as I go.

Please help me understand this correctly.

Other file

The Plc Kid said:

Just for educational purposes what if I had variable voltage/phase angle would that effect the accuracy of my ct with the ct I am using . This model says sinusoidal and non sinusoidal currents.

I think I am getting more lost as I go.

Please help me understand this correctly.

Click to expand...

Your Din-a-mite file didn't work, but I am familiar with them. Yes, I believe those CTs would work with phase angle control. The key is the non-sinusoidal spec. You need this with VFDs, SCR power controllers, just about anything with solid-state control.

Brian

I think the start of your confusion is your management is not clearly laying out what they want.

Phase angle applies only to loads with reactance ie inductors (motor windings, transformers lighting ballasts etc) or capacitance. Heaters are generally thought of as resistive only - EXCEPT inductive heaters. I assume you have only resistive heaters - therefore PF is unity and current is in phase with voltage.

What you do to one side of a transformer you do to other side P = P. Step up or down volt and current are inversely proportional from pri to secondary. Phase angle on a single CT should not matter. If 3 phase and they are interconnected on secondary maybe so especially if one phase PF differes greatly from others.

I think the big issue for you to decide is four fold
1. what management wants
2. Power or current
3. Individual heaters or the bank
4. where to take measurement from controller side or from output side

As you have mentioned if these are just switch controlled heaters (and not staged) ie 0 or 100% off or on it is simple.
%power = actual / theor (or total measured). A simple ammeter with a green band would suffice for this criteria and setup.

With PWM control the power output varies so you have two choices
%power = actual / total connected (erroneous at low levels)
%power = actual / what PWM is supposed to deliver at this point in time. This is the tougher of the two and I am sure I am not expressing this well as I could/should. I think you would divide actual power by the error between setpoint and process variable of your controller and then throw on a fudge factor to obtain a simple %power reading. This value I guess could be called
"percent of demanded power delivered"
OR delivered vs demanded power ??
and with all heaters OK should always read 100%.

OK gurus how about some help here?

After you get this all done just for grins and giggles you can setup an algorythm for heatup rate ie dT/dt and then compare that to desired rate. When you are done with that I'll have some other little project for you.
Dan Bentler

You need hard data. Good news is you have access to it. Use the current sensors you have to answer the question "Would the heater pull the same amount of current at a cold platen state as they would if it were in 2-3 degrees of setpoint?".

For the on/off controls, the approach you outlined "I guess on the on/off systems I could do logic that if the heat is on for that zone and the current being pulled is below limit then set an alert." should work fine if the answer to the question above indicates it doesn't vary too widely. It's also going to matter how many heaters are in the zone. If your zone only has 4 heaters, detecting a 25% drop should be easy. If it has 20 heaters, a 5% drop may be impossible to detect. Especially if the startup current turns out to be 20% higher than the "at sp" current.

Likewise, for the modulated controls, if you can coorelate the actual current to the %output, you can check the difference, and if it exceeds a set % of output, set an alarm. If when you trend %output and actual current, they do not track well, then you have a pretty impossible task.

So, go get your data. Show it to the people asking you to add this feature and come to a conclusion about how useful this will actually be. If it works, it will be great. But if it just generates nuisance alarms, it's a waste. You should have a good idea how risky it is once you have some good data.