CraftWEB Hot Glass Talk

CraftWEB Hot Glass Talk (
-   General Hot Glass Discussion (
-   -   Useless controler mypia ta4 (

Franklin Sankar 07-08-2020 04:15 PM

Useless controler mypia ta4
I canít believe I bought a controller for my furnace and canít find any programming to ramp it up.
What a waste. I canít believe they sell a controler you canít ramp up.

Larry Cazes 07-08-2020 06:20 PM


Originally Posted by Franklin Sankar (Post 148158)
I canít believe I bought a controller for my furnace and canít find any programming to ramp it up.
What a waste. I canít believe they sell a controler you canít ramp up.

just fixed setpoints?

Pete VanderLaan 07-08-2020 06:41 PM


Originally Posted by Franklin Sankar (Post 148158)
I can’t believe I bought a controller for my furnace and can’t find any programming to ramp it up.
What a waste. I can’t believe they sell a controler you can’t ramp up.

Well indeed they do. Ramp and soak has gotten more, not less expensive over time. Read the fine print, and then, read it again. I have multiple useless controllers here from Dwyer that do set point and no more. They have their uses but rarely do things a person wanting to cool products slowly would want. In LOVE, the 16 A is miles from the 16 B.

Franklin Sankar 07-09-2020 08:21 AM

Auber seem to be cheap and nice features.


Randy Kaltenbach 07-09-2020 09:00 AM

Last year I bought a cheap ramp / soak controller from Thermomart in Canada. About $100 CDN

So far, so good. It's been through about 8 cycles running a kiln.

The documentation is marginal, but it was enough for me.

Franklin Sankar 07-09-2020 09:29 AM

New product release! USB ramp soak PID controller SYL-53X2P! Monitor and control your PID via your computer!
Looks good for 72$

Pete VanderLaan 07-09-2020 12:07 PM


Originally Posted by Franklin Sankar (Post 148163)
Auber seem to be cheap and nice features.


We have two aubers and the method of doing set points drives me nuts. I can't stand it. I tore it out.

Franklin Sankar 07-09-2020 01:19 PM

Yes they have a jump instruction that really makes you jump.

Eben Horton 07-09-2020 05:10 PM

Pete V has a love controller on his dishwasher annealer. Holy crap ... operating that thing is like trying to decipher the Dead Sea scrolls.

Check eBay For a used Digitry gb1. They are so easy to use.

Pete VanderLaan 07-10-2020 07:46 AM

Eric retired the dishwasher after several decades and that controller is now on the new one he built me. I agree that loves are quirky but I don't like the notion of one controller running more than one tool.

Shawn Everette 07-10-2020 10:12 AM

I too am not a fan of the Auber. I got a plug n play step controller, but it's logic offset is way off. I have to start it 40* higher than I want it, then slowly dial it back down til it settles out in about 2 hours. Luckily it's for a flame kiln, so flexibility of boro.

Eben Horton 07-10-2020 10:48 AM


Originally Posted by Pete VanderLaan (Post 148173)
Eric retired the dishwasher after several decades and that controller is now on the new one he built me. I agree that loves are quirky but I don't like the notion of one controller running more than one tool.

Thatís where the GB 1 comes in. It only runs 1

Franklin Sankar 07-10-2020 02:42 PM

Shawn how did you know you were 40 deg off from the actual.

Shawn Everette 07-11-2020 08:33 AM

Doesn't go to set temp, gets within 40* and just slowly floats up. Not a power or insulation issues, target is 500* below max temp.

Franklin Sankar 07-12-2020 02:27 PM

Oh. I see what you mean. Did you do auto tune?

Shawn Everette 07-13-2020 04:09 PM

This particular model isn't their standard PID. I've done an auto tune, but still have to mess with some of the parameters. Right now it's not at a point where it's causing problems, so I just deal with it.

Pete VanderLaan 07-14-2020 10:51 AM

Do a slump test for calibrating.

Franklin Sankar 07-14-2020 04:39 PM

Nice strategy Pete. I am using mine on the furnace.

Shawn Everette 07-15-2020 12:10 PM

That's not where my issues is. When I start out I have to set it to 1080 for it to settle on 1050. As the day goes on I slowly lower the temp, and after about 2 hours I can have it set at 1050 and it'll stay there. It's not a thermo error, but a controller logic one.

Franklin Sankar 07-16-2020 02:06 PM

I am having a bad controller time.
I have a Watlow 93 and set it on calibrate inadvertently.
Itís bad. I tried to stop it but too late. It went through the sequence and The thermocouple started to act up. Showing ctazy numbers. I recalibrated again and restored the factory settings. it seem to work. But I know something must be wrong because one of the steps is to calibrate the thermocouple using a j type stub in ice which is 0;deg. I donít have that. So My temp reading is off badly.
Thanks Shawn for sharing your offset trick. Hope I can apply it now.
Thinking about Using another controller to measure the temp and comparing both readings.

Franklin Sankar 07-16-2020 02:34 PM

My actual value is lower than the set value Of 2000 by about 40 deg at steady-state. It will never reach 2000
So if I have a set value of 2000 the controller will show a reading of 1860.
If I add offset of say 40 the actual value reading will show 1860 +40
But the Is actual temp 1860 or 2000.

Brian Wong Shui 07-16-2020 11:57 PM


You can check the readings by putting the thermocouple into a cup of melting ice (lots of ice with water surrounding the ice) and the controller should read 0C (32F) and into boiling water 100C (212F). This should give you an idea of if the calibration is way off. If you restored factory settings, it probably isn't too far off.

Have you tuned the controller? When tuning, you need to tune around the expected setpoint. i.e. don't autotune at 200F and expect the tuning to hold at 2000F.

I've typically tuned this generation of watlow controllers manually because the autotune makes the output hyperactive. They have a decent procedure on manual tuning in the controller manual

A 40F steady state droop seems to me that the PID settings is mostly Proportional Band and you might need a little Integral (Reset) to close the error. The Derivative will affect how fast the system gets back on the gas back to the set point after a when you open the lid. Too much makes is really slow to get back to the setpoint. Too little makes it overshoot the setpoint.

Franklin Sankar 07-17-2020 05:26 AM

Thanks Brian. Lots of good stuff to try.

Franklin Sankar 07-18-2020 07:16 PM

I fixed the difference in temperature Between the set point and actual by tuning it but I have a slow cycling about 3 deg . Canít get steady state .
Pb1 at 1, if I go higher it will stead state at about 25 Deg or morE from the set point.
So I wil live withy the small cycle. Maybe play some more.

Franklin Sankar 07-18-2020 07:25 PM

P = 1, I = .01 , D= 1.

Brian Wong Shui 07-19-2020 07:18 AM

Hi Franklin

The Numbers looks a little off with Pb looking too small for a furnace. Here is a procedure from watlow that Iíve used to manually tune.

Franklin Sankar 07-19-2020 09:47 PM

Thanks Brian. I was using the following which I found in the watlow 93 site
I will try yours tomorrow.

1. Apply power to the Series 93 and enter a set point. Set Operation parame- ters as follows: [``Pb] to [```1], [``rE] / [``It] to [`)00], [``rA] / [``dE] to [`)00], [``Ct] to [``%0], [`CAL] to [```0], [`AUt] to [```0].
2. Proportional Band Adjustment: Gradually increase [``Pb] until the upper display temperature stabilizes to a constant value. The process tem- perature will not be right on set point because the initial reset value is 0.00 repeats per minute. (When [``Pb] is set to 0; [``rE] / [``It] and [``rA] / [``dE] are inoperative, and the Series 93 functions as a simple on-off con- troller.) The [`HSC] parameter determines the switching differential value.
3. Reset/Integral Adjustment: Gradually increase [``rE], or decrease [``It] until the upper display temperature begins to oscillate or "hunt." Then slowly decrease [``rE] or increase [``It] until the upper display sta- bilizes again near the set point.
4. Cycle Time Adjustment: Set [``Ct] as required. Faster cycle times sometimes achieve the best system control. However, if a mechanical contac- tor or solenoid is switching power to the load, a longer cycle time may be de- sirable to minimize wear on the mechanical components. Experiment until the cycle time is consistent with the quality of control you want. [``Ct] will not appear on units with a process output.
5. Rate/Derivative Adjustment: Increase [``rA] / [``dE] to 1.00 minute. Then raise the set point by 11į to 17įC, or 20į to 30įF. Observe the system's approach to the set point. If the load temperature overshoots the set point, in- crease [``rA] / [``dE] to 2.00 minutes.
Raise the set point by 11 to 17įC, or 20 to 30įF and watch the approach to the new set point. If you increase [``rA] / [``dE] too much, the approach to the set point is very sluggish. Repeat as necessary until the system rises to the new set point without overshooting or approaching the set point too slowly.

Franklin Sankar 07-20-2020 10:21 AM

I tried auto tune one more time and it worked.
I dropped 200 from the set point then started the auto tune and it worked like magic. I feel the next time I start up the settings will not work at start up temp but will do a work around if that happens.
I will note the parameters used so next time I will be ok.
Thanks for all the help

Franklin Sankar 07-20-2020 10:31 AM

P= 7 I=.1 d=.47
It will be interesting to see what happens next startup.

Brian Wong Shui 07-21-2020 08:42 PM

Thatís great.

Next time the furnace is up, check to see how the furnace behaves when you upset it from steady state (i.e opening and closing the door when gathering glass). Look at output percentage. As soon as it detects the temperature is falling, you should see the output jump to 100% right then. It should overshoot the setpoint and then fallback to slightly under and then converge on the setpoint. The output percentage should be fairly steady once it finds the setpoint and not erratically jumping around.

If it takes too long to get back, try reducing the d a little. (You did record the values, right?). Too much d makes the system slow. Ideally you want the system slightly underdamped to critically damped. This will have the fastest response back to the setpoint.

Franklin Sankar 07-22-2020 09:46 AM

Thanks will do. Values recorded above

All times are GMT -5. The time now is 08:24 AM.

Powered by vBulletin® Version 3.7.2
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.
© Opportunity Network. 2008. All Rights Reserved.