View Full Version : Technical Kiln & Controller Help Needed
12-21-2006, 02:10 PM
Hello, I'm hoping someone with some electrical knowledge and/or kiln or controller building experience can help me out.
I have an Aim brick kiln with an infinite switch and I have built a controller for it using a Fuji "brain". I leave the kiln infinite switch on max and I expect the controller to do the heat adjustment as needed.
The controller works fine most of the time, but when the kiln gets pretty hot (like 900), sometimes the kiln continues to heat way past the setpoint even though the controller is signaling to turn the power off - it just keeps going and the power does not cycle off. The kiln will continue to heat forever. It's as if the kiln "locks" the power and just keeps the circuit open and keeps draws power even though the Fuji is signaling to stop the power (and the Fuji IS signaling to stop the power, I've checked it). Also, when I select the "Autotune" function on the Fuji, it almost always locks the power on sooner or more readily than when the autotune process is not running. Not sure if that matters or not. Now, here's a twist. If I turn the infinite switch on the kiln down to maybe 5.5 or 6 (out of 10), the kiln and controller seem to operate fine, but of course the temperature will never get hot enough for my use since the power going to the kiln is being attenuated by the infinite switch. Odd, huh?
Here is an experient I did. I connected my kiln and a small test lightbulb to the controller at the same time with a plug splitter. When the controller sends power, the light bulb lights up and the kiln also receives power. When the power "locks" on and the kiln continues to heat past the setpoint, the test light stays on with the kiln, as you would expect. Then - here's the important part - when I turn the kiln off with the infinite switch so it draws no power, the test light begins to turn off in about 1.5 seconds and fades to black over about 1 second. It does not go black instantly as if you flipped a switch in your house. So when the kiln draws no power through the controller, the circuit breaks as it should. That's why I use the terminology that the kiln "locks the power on".
Here is another weird finding. If I reach behind my controller box and turn the controller off, so that the fuji screen is totally dark. That test light I mentioned tries to light up. It is very very dim, barely noticable, but it has a flicker of power about 2 times per second in a consistent evenly spaced rythmic pattern. I'm not sure if that is a normal power analysis "ping" from the Fuji or if that should not be happening or what it might indicate. Again, the fuji screen is dark, and the controller is "off" at the time.
I am using a solid state relay in the controller and the symptoms indicate to me that the relay is probably the problem. It seems to me that the relay is not closing the ciruit when instructed by the Fuji because of the high power being drawn through it by the kiln and is probably not behaving as it was designed.
Does what I'm saying make any sense? Any ideas on what might be going on or suggestions on what to do? Also, do any of you have any advice on who I might PM or talk to about this. Who could help me? Thank you! :wave:
12-21-2006, 06:52 PM
Well... you have taken a very logical approach. I undertand the problem is the kiln continues to heats when it should not?
The only thing that can give power to the kiln is the controller (which is a PID and a SSR). If you are certain the PID is signaling to stop power... then the SSR is not responding, and it is giving the kiln power. You did not say how you verified the PID was sending a 'off' signal to the SSR, but we will assume the PID is working as expected.
There are only 3 possibilities, and as you suspected they are all 'relay', i.e., SSR oriented.
1. The SSR is wired to the wrong terminals on the PID. You may be driving the PID with the wrong variable voltage solid state output. Not likely... but could happen. If so... see PID manual... and rewire if needed.
2. The SSR is mismatched for the PID. Possible... more likely... but not most likely. Note the SS drive currents in the PID manual and determine if the SSR matches. If not... replace SSR, and see next item.
3. The SSR gets too hot and becomes erratic. Most likely the problem. SSRs are not mechanical relays... they can get too hot if they don't have a heat sink, and if the amperage draw is too high. The SSR should be rated to carry the amperage required to operate the kiln. If not... replace SSR with properly amperage rated SSR -- with a heat sink attached.
As for everything else... the PIDs use extremely complicated algorithyms that have no resemblance to common sense. In some modes the will go off into neverland and borg the system... to 'learn' the system they are controlling... even allowing the heat to climb way past the set point... until they get what they want to know. It can be frustrating programming a component without the actual source code to know what it really does. Suggestion: Pick the simplest routine and use it exclusively.
You might do a little googling to learn how SSRs work, if your interested. They are strange little beasts.
Real men use mechanical relays :) .
12-21-2006, 08:31 PM
Hey thanks Padre. I think you have the solution after I have talked with another person regarding the problem. I did not mount the SSR which I am using to a heatsink and I think that is causing the SSR to behave eratically due to the heat.
It is also in a project box from radio shack which has no ventilation holes in it. I plan on buying a heatsink and giving the relay some "air conditioning" and see if the problem goes away.
I would like to use a mechanical relay, but I need one that actuates or charges the coil with less than 20 milliamps, which I cannot seem to find.
I'll follow up with you and let you know what the problem was.
By the way, I DO have one EMR in the project, which ensures that if there is a power failure and the power then comes back on, the kiln will not resume operation. So I'm not a total wuss. http://www.lampworketc.com/forums/images/smilies/wink.gif
12-21-2006, 11:14 PM
If your Fuji controller is like mine, you don't want a mechanical relay instead of the SSR. My controller sends a signal every 2 seconds to the relay.
If you ran your controller for 8 hours the relay would cycle 14,400 times. You would exceed the specs of a traditional mechanical relay in a couple of weeks.
I understand EMRs would be instead of the SSR too, although I have never used one. Other than cost, and an upright orientation, I can't imagine a problem withthem. You might run it by your supplier, just in case the controller would not work with it.
12-21-2006, 11:27 PM
There is another possibility to consider.
Is your thermocouple pushed too deep into
your kiln causing a direct short?
That would make it want to call for heat also.
Also you might consider using a mecury relay
as their cycle lifetime is rather long.
12-21-2006, 11:57 PM
Dan it could also be the way you have the parameters set up,what fuji do you have PXR3 ?if so.
first thing you need to is get rid of the infinite switch I have no idia why people want to run a controller through them,come right off the relay to the element.If you have left a few of the factory settings on the perameters this could be why you get alot of bounce.hell just call me I will hook you up.
O yeh and I have no clue why you are using a coil,you can plug all the fuji's right into 110 volts or 220.
12-22-2006, 11:20 AM
I will check for a direct short with the thermocouple, but I doubt that is an issue. Like I said, everything works great at lower power levels. Thanks for the idea though.
I do have a PXR3 controller. I would like to get rid of the infinite switch as you suggested, but to do that would void my kiln warranty which I would like to avoid. I tried to order the kiln without an inifinite switch and I could not do so, they didn't offer it.
I've checked the parameters pretty thoroughly and I think they're okay. The main problem I'm having is not really bounce, it's up, up and more up. It never turns off once it gets to a higher temp like 900 - just through the roof.
O yeh and I have no clue why you are using a coil,you can plug all the fuji's right into 110 volts or 220.
I'm not sure what you mean here. The only coils I know of are the heating coils inside the kiln and the coil in an EMR which is not the relay I'm having the problem with. My problem is with a solid state relay.
Thanks for the help guys!
12-26-2006, 10:53 AM
i saw you posted this at GT as well. in my temperature controller tutorial, i am sure i mentioned some of the issues with solid state relays (SSR). on the good side, SSR's provide better temperature control, longer life heating elements, and long life. but SSR's consume power, and thus they get hot, so they need a big heat sink and a fan to keep them cool. the box needs to have slots in it and the box should be metal, preferably aluminum, but definitely not plastic. SSR's also tend to fail ON so that if they fail, your kiln will go into thermal runaway.
for these reasons, i recomend that people use electomechanical relays (EMR) for most applications. very little power is actually consumed by the relay so they don't get hot. therefore you do not need a heat sink, or a fan, or a metal box with slots in it. they are very low cost, and easy to service. EMR's tend to fail OFF so your kiln will shut down.
a SSR normally turns on/off every 0.5 to 2 times per second. but a EMR cannot operate that fast, and normally has a 20 to 30 second cycle time. temperature controllers usually have an SSR output, or an EMR output and some controllers offer both. you need to check what type of output you have. if you need help, post the EXACT model number and i will offer assistance.
12-27-2006, 11:15 AM
Hi Mark, thanks for the help. Here are some specifics on my setup.
The SS relay is an Omron G3NA-2258 and it is marked as follows:
24-240 vac / 25 amp load
5-24 dc input
My kiln should be drawing 14-15 amps.
I attached 6 small brass heatsinks to the metal plate on the back of the relay, and butted them up against one another for better heat distribution, then I ran the kiln again. Same result. http://www.lampworketc.com/forums/images/smilies/cry.gif
But, when I put my hand on the heatsinks while it was running, they were HOT. Damn hot. Too hot to leave my hand on it for even a second without getting burned. The relay that I have in that setup should not be getting THAT hot should it? I mean, I know it's going to get "hot" but that was really hot.
My controller has several sets of numbers and I'm not sure which one is the model number but here they are anyway:
1) Marked on a label on the controller itself is the number "PXR3RCY1-4VOA1".
2) Marked on the same label is "No: 5Y24026T". I think that is a serial number.
3) On the paper specs/instruction sheet that came with the controller is printed "INP-TN1PXR3d-E". Note the lower-case "d".
Thanks again! :wave:
12-27-2006, 12:31 PM
Here's the PID specs:
PXR3-RCY1-4V0A1 3 - 1/32 DIN (24x48mm) R - Thermocouple (°F) C - SSR or SSC drive output Y - None 4 - None V - Standard (100-240VAC, 50/60Hz) 0 - None -- per: http://www.instrumart.com/AllPartNumbers.aspx?ProductID=4714
It is a basic PID, with Thermocouple (°F) input, and SSR or SSC drive output.
Here's a photo of my SSC (Watlow). The black heat sink behind the black SSC (back-to-back) is aluminum and is 2X the size of the SSC - and it's only 18A.
I've never opened the box up when running so I have never touched the SSC heat sink to see how hot it is . The grey enclosure box gets warm.
Perhaps Mark has opened his and touched it... he's pretty brave :) .
Or Mike C.
p.s. Don't tell anyone that one of my controllers is a non-EMR. Real men use mechanical relays. All joking aside... I really do prefer EMR for glasswork.
12-27-2006, 01:11 PM
Ok, so I've gotta use an SSR with that controller, right? I guess I'll just heatsink the heck out of it and see if it will work for me. If not, it must be defective. Thanks!
12-27-2006, 01:40 PM
Yes, that configuration requires a SSR. Mine get hot, but not like you are describing. My only heat sink is some heat-sink grease on the aluminum bottom to my Radio Shack box.
12-27-2006, 01:48 PM
Steve, are you saying that your heatsink is stuck to the aluminum bottom of your radioshack box with heatsink grease compound and nothing else?
Also, do you have the plastic coverplate attached over (i.e. on top of) the aluminum bottom or is the aluminum exposed to the air on the bottom of the box?
Finally, what kind of SSR are you using?
12-27-2006, 02:01 PM
I have the SSR bolted down to the aluminum bottom, with the grease in between for good contact. I have the cover on the Radio Shack box in place. It has rubber feet that keep the aluminum bottom about 1/4 inch off of the table.
I have no recollection what brand SSR I am using. I got it from TTI Global. I used the one in the list of materials from Toblers site:
My controller looks just like the one in the photo, except I went with a different thermocouple wire setup.
12-28-2006, 08:04 PM
I run a Omron E5EX-A controler with a 1 amp SSR output driving an external 120volt input/120 volt 25 amp output SSR. I found that I had to put a 100K ohm resistor across the output of the controller to turn off the big SSR. The little SSR in the controller is unable to bleed off the control voltage of the bigger SSR. Your main SSR is 5-24 volt DC input,the controller is 16vdc max, how are you suppling the DC volts?
For a heatsink, a CPU heatsink from an old computer left at our local electronics recycle drop off does the trick.
12-29-2006, 02:07 PM
Randomthoughts, I'm not sure I understand your question.:confused: The controller that I'm using supplies about 12 vdc to the SSR. The SSR will complete a circuit when fed between 5 and 24 vdc. The circuit it completes is a household 120v, which drives the kiln.
12-29-2006, 03:48 PM
The omron SSR's do not supply voltage, they are a relay. with no wires connected they will float to a voltage, but do not consistently supply one, therefore an external voltage is required, (the internal supply is a different option). I note that the fuji manual calls it's output SSR/SSC, so this might not apply in your case. However, if it does want an external supply, and you are trying to control with just the voltage that bleeds through the electronics, it would explain your problem. If you connect a small light (or LED and resistor) to the output in parallel with the input of your Big SCR, it will either a) bleed off any float voltage and the light will never light/SCR will never fire, or b) show you that you have a good input to the main SSR.
Another way this problem can happen is the controller's SSR just stops supplying voltage to the main SSR, but does not bleed off what is already there. If the leakage currents in the main SSR at high power levels are high enough, they will maintain the input. A bleeder light or resistor will take care of this type of problem.
The symptoms of the test light fading to black over seconds is compatible with voltage bleeding through from/to high impedance inputs and outputs. This is why I have to have a bleed resistor in parallel with my controller's SSR's output, without it the main SSR never turns off.
Not a guaranteed fix but you have covered most of the other obvious problems
01-02-2007, 10:53 AM
Update for those of you who answered. Just thought I would let you know that I have solved the problem. It was definitely that the SSR was overheating and malfunctioning. I got a sheet of 1-2mm aluminum that I had in the kitchen. It was originally sold as a "thawing tray" for frozen food. You put the food on it and it thaws it very quickly. It's just a giant heatsink.
Well I clamped the SSR onto it with some thermal compund in between and it runs like a champ. The aluminum sheet is about 6 times bigger than the SSR, which I hope is overkill and I can cut it down some.
So, the moral of the story is, keep those SSRs COOOL!
01-02-2007, 10:59 AM
Thanks for the update! Now we have a good data point for others who might have a similar problem. Appreciate it:clap:
vBulletin® v3.5.8, Copyright ©2000-2014, Jelsoft Enterprises Ltd.