Red Lion Modular Controller Retrofit

Hello,

I know we have a couple people on this forum familiar with the Red Lion product line.

I have a sealing machine in the plant currently using a Red Lion modular controller with two CSPID2 temp controllers (relay output). It is controlling four 250W heaters.

(see http://www.redlion.net/Products/Groups/ModularControllerModules/CSPID/Docs/05023.pdf)

Since the power to the heaters is switching on/off the PID loops tune in to adjust the switching time/frequency. The temperature is held to +/- 2 degrees F or so.

Although this was initially good enough, I am now finding that there would be benefits to going to a solid state controller that can switch much faster than mechanical relays that I currently am using. This should allow me to control the temperature to +/- a tenth of a degree or so I figure, since there is a large heat mass and the temparature swings very slowly (as opposed to heating a large airflow for example)

Two questions:

1. I am looking at the spec sheets and I am confused as to whether or not I need the 'solid state' or the 'triac' output. I am thinking it is the triac, but I am not positive on this.

2. I should simply use the output from the module to energize/de-energize separate solid state relays wired to the heaters, correct? My peak load should be around 2A given the power/voltage used, which is too high to switch directly.

If someone could please help me out, I would appreciate it greatly.

Thanks!

DL06,

Some SSRs are triacs that are opto coupled.
Try to Google the spec sheets.

Rod (CNC dude)

Hi Rod,

Forgive my inexperince, I am not sure that I completely follow your posting.

You are saying that I need a triac version of the CSPID connected to an 'opto coupled SSR' which is in turn tied into the heater circuit?

Thanks

I will do some more reading/learning on SSRs. I was thinking that a Automation Direct AD-SSR210-AC would do the trick.

DL06,

I don't have any experience in what you are attempting.
Someone else will know.

However, if you have heater coils that are 120 VAC, one traic SSR properly amp rated would drive the coil. But I imagine you have 240 VAC coils.

I really posted to keep this up for viewing.

Rod

Thanks Rod. The heaters are actually 120 VAC, which I meant to have stated more explicitly (I only said 250W and ~2A).

If you use the triac output to drive the triac input of the SSR you have mentioned you might get turn-off problems. You might have to use a swamping resistor to ensure minimum load on the controller output.

I'm not sure you'll get tighter control making a hardware change.

Most systems like Red Lion have a setting called something like 'cycle time', which is the

period of time in which the output switches on and off in proportion to the PID calculation,

where 100% is full on, 50% is on half the time, off half the time, and 0% is off all the time.

Sometimes the cycle time is limited to some minimum value for relay outputs to avoid shaking

the relays to death. Sometimes the cycle time can be reduced to fractions of a second when

there's a solid state output that won't wear out like electromechanical relays do.

What is the available minimum cycle time? Can it be reduced lower than your current setting?

If you're not getting as tight control as you want, it could be more of a tuning issue than a cycle time issue.

You should check and make sure that you can acutally reduce the cycle time.

If you decide to change, yes, you want to get a solid state relay to drive your heater elements.

Make sure you heat sink it well and it'll last forever. There are two kinds, one with an
AC input (coil) and the other with a DC input (coil) which are normally rated 3-23Vdc input

driver circuit.

I'm sure Red Lion has some sort of solid state output for its system that matches either style, whether AC or 3-32Vdc.


Dan

Personally I do not see an issue with the system as is...looks to me like someone just wants to play with something.

Lets take a low standard heat value..ie 350 degrees, +/- 2 degrees is nothing...less that 1/2 a percent. Make the temp setting higher..ie 800 degrees and +/- 2 is .25%

As far as I can tell this is a digitally controlled system and it does not matter how fast you can switch devices but how good you can maintain the temperature setpoint...to me .5 of 1% is good regulation.

If you're not getting as tight control as you want, it could be more of a tuning issue than a cycle time issue.

Click to expand...

How can you tell the difference if poor tuning causes the 2 degrees of error or the switching isn't fast enough?

Lets take a low standard heat value..ie 350 degrees, +/- 2 degrees is nothing...less that 1/2 a percent. Make the temp setting higher..ie 800 degrees and +/- 2 is .25%

Click to expand...

It looks even better if you measure the temperature in degrees R. Think about it.

I assume you refer to Degrees Reaumur , 1 being 1.25C .



1% is good regulation ? just glad I don't ever have to fly in it .

No Rankine

http://en.wikipedia.org/wiki/Rankine
but does it really make a difference? I was just making a subtle point that one must specify a range before specifying a percentange. Is 2 degrees out of 800 equal to 0.25%? Why not 2 degrees out of 800+459.67 or 0.159%?

Then the difference is range only as you state .

The working range of the process is what is important , the mistake made in a number of statements is where a poster states that 2 degrees in 350 degrees (lets stick with F for now ?!) is 0.5% , how so ? what is 350 F , relative to absolute zero ? I don't see how you can state a degree of error as a percentage if you don't know what 100% is .
Surely percentage of error doesn't change dependent of measurement units ?

Sorry for the delay in checking back - thanks for all of the input guys.

To answer Danw - I believe that the triac can turn on and off many times faster than a mechancal relay. From my understanding (someone please jump in if I'm totaly off base), a triac can turn on at a certain phase of the AC sine wave half cycle (120 of which are present each second). So the modulation can be incredibly precise compared to relay based on/off control.

Regarding the accuracy issue, greater accuracy would increase the seal produced and decrease the number of detects/rework/etc. As long as there are defects, I am never okay with saying that it is 'good enough' as it is against the philosophy of lean manufacturing, six sigma, continuous improvement, etc. The four degree swing does cause problems, so why not fix the problem? True, the machine operates at about a 1% reject rate. To some, this sounds great, to those practicing six sigma it sounds like a great opportunity for improvement.

I am heating a fairly large aluminum sealing head, so I just would have thought that tighter control would be possible with a triac unit.

Okay, let me flip this around now. How about if I use a smaler heater so the temperature swing is minimized when the heater turns on and off, while not increasing the minimum cycling time and putting more wear and tear on the relays?

Thanks again for the discussion gents, great information.

You have several questions on the table.

A Red Lion relay output will always have a recommended slower cycle time in order to preserve the life of the relay. Even with that, the relay will wear out. (10 second recommended cycle time x 24 hours/day x 365 days = millions of cycles/year)

Using the Red Lion triac output permits a faster cycle time. That is a good choice for activating “AC control input” solid state relays. The triac will not wear out. It does have some leakage. Either: put a burden resistor on the input to the solid state relay OR use a solid state relay that already is designed with this in mind (such as www.power-io.com or certain www.omega.com brands.) Set your cycle time to a fast number.

Or use the Red Lion solid state output. This is a good choice for activating “DC control input” solid state relays. Nothing will wear out. Set your cycle time to a fast number.

A fast cycle time (like one second or less) will offer high stability and fast response.

If your application typically stabilizes somewhere between 25-75% output to maintain temperature, then your heaters are probably sized correctly. You have a large aluminum sealing bar. You want to use a fast cycle time and do NOT undersize the heaters since you want “super fast” recovery on the sealing surface to prevent bad heat seals.

Answer The Question!! Stop Guessing!!!

How can you tell the difference if poor tuning causes the 2 degrees of error or the switching isn't fast enough?

Click to expand...

If you turn the power off, how much does the temperature decrease in one time period? If you turn the heater on, how much does the temperature increase in one time period. If the sample period is reduced by half the temperature will change roughly by 1/2 in one time period. However, if the temperature doesn't change much in one time period then doubling the sample rate will not make much difference.

One should also be aware that the system will heat up much faster when at the low of of the temperature range than at the top.

All this can be calculated if you know the time constant of the system.

DL06 said:

I am heating a fairly large aluminum sealing head, so I just would have thought that tighter control would be possible with a triac unit.

Click to expand...

SO ARE YOU GUESSING? The point is that the aluminum bar and heater have a time constant ( how do you calculate that? ) and if the sample period is 10 faster relative to the time constant of the alunimum bar then making the sample time faster will not make much difference. Do the math. Oh, what math? What is the formula that has to do with sample periods and time constants? I have shown the forum many times and you all ignore it. You all prefer to guess when this can all be caluclated. If faster sampling is required then how much faster?