[/quote]Which harks back to the thread we had a few years ago: "SSR failure: The China Syndrome"
Firing Un-Attended & Failure of Relays
Moderators: Brad Walker, Tony Smith
-
David Williams
By the way, Three Mile Island reached a temperature of 4300 F, so the elements would burn out long before the kiln did a Three Mile Island. That's not to suggest that a kiln's a great place to store your excess uranium, of course. [/quote]
Which harks back to the thread we had a few years ago: "SSR failure: The China Syndrome"
[/quote]Which harks back to the thread we had a few years ago: "SSR failure: The China Syndrome"
As usual COMMON SENSE RULES. Set it up right and sleep well. The more expensive is your control equipment, the more reliable it should be, I hope. (Brock's controller is the most expensive but only he knows just how reliable it really is)
It's still more reliable than a manufacturer mounted controller. Just had another one go in a class. It was 4' by 8' Denver, and during a slump firing, a mercury relay went, and the kiln got up to 1747F for some time. The controller was supposed to have an upper limit of 1600, so 2 things failed. Did you know you could slump 3/16" plate steel molds? Well, you can! They had been rolled into a gentle arc, (about 1 1/2" deep over a 22" span) to use as a sushi mold, and now they are flat again.
BUT, we may have inadvertently discovered a new twist on working hot. Make a random pattern bar, slice it, and re-arrange it, then boil it. Verrry interesssting. The book-ended Rohrschachian patterns bleed out and vary in unpredictable ways.
Brock
It's still more reliable than a manufacturer mounted controller. Just had another one go in a class. It was 4' by 8' Denver, and during a slump firing, a mercury relay went, and the kiln got up to 1747F for some time. The controller was supposed to have an upper limit of 1600, so 2 things failed. Did you know you could slump 3/16" plate steel molds? Well, you can! They had been rolled into a gentle arc, (about 1 1/2" deep over a 22" span) to use as a sushi mold, and now they are flat again.
BUT, we may have inadvertently discovered a new twist on working hot. Make a random pattern bar, slice it, and re-arrange it, then boil it. Verrry interesssting. The book-ended Rohrschachian patterns bleed out and vary in unpredictable ways.
Brock
My memory is so good, I can't remember the last time I forgot something . . .
-
David Williams
The controller was supposed to have an upper limit of 1600, so 2 things failed.
You can set the upper limit of the controller; if the relay fails you will still have a runaway. The controller was blameless. It should have given you an error reading though.
I think if people are really worried about it, overtemp systems are easy and cheap. I worry way more about the propane refrigerator in my home that I leave running while I'm doing shows, than I do about my kilns. Its unattended open flame that makes me nervous. The main danger with a runaway fusing kiln is like I said if you have a decent volume of glass it will bore through the soft brick and could conceivably hit the ground. So if you had the kiln on a wood floor or carpet...
You can set the upper limit of the controller; if the relay fails you will still have a runaway. The controller was blameless. It should have given you an error reading though.
I think if people are really worried about it, overtemp systems are easy and cheap. I worry way more about the propane refrigerator in my home that I leave running while I'm doing shows, than I do about my kilns. Its unattended open flame that makes me nervous. The main danger with a runaway fusing kiln is like I said if you have a decent volume of glass it will bore through the soft brick and could conceivably hit the ground. So if you had the kiln on a wood floor or carpet...
-
Phil Hoppes
- Posts: 298
- Joined: Wed Mar 12, 2003 2:20 pm
- Location: Overgaard, AZ
Root cause analysis
Brock,
Do you know for a fact that the merc relay failed or was it in fact the control signal to the relay that stuck on? It is my understanding that a merc relay turns off from gravity thus cannot fail in the on position. The only way it could fail on is that the enegerizing coil was on thus the failure was actually one step if front of the relay not the relay itself.
I've had a number of side discussions with some people on this topic as I'm working on my kiln design as I speak. True, as all have said, if your kiln is placed in a safe envrionment (no flamables close by, good ventilation, etc.) the worse that can happen is that you have a puddle of glass on the floor and some damaged insulation along with a nice power bill. That being said I really don't wish that to happen. Also, I wonder as I mention above, how much "root cause" analysis is done wrt the actual failure mechanism. I reversed engineered my Paragon control electronics on my GL24. For starters the "real" schematic is nothing like what is published on the Paragon web page. That schematic is a joke. Looking over the real schematic there are 3 main failure points (and some smaller ones) that could cause one or more elements to "run away". My point is, that when a kiln goes bonkers there could have been a number of reasons for the event. Sometimes, especially if the cause is an intermittent failure of a component, it is very difficult to determine the cause as the problem is not repeatable. There is also one other cause that has not been mentioned which is plain operator error. Nothing broke, the controller did exactly what it was told to do, the operator simply told it the wrong thing. I am mildly dislexic, especially with numbers. It would be real easy for me to slip a digit here or there. I usually double check my programs after I have entered them because of this.
This is an interesting problem. Both from an analysis and feedback point I would be interested in hearing firsthand from anyone out there who has had/seen actual kiln meltdowns and in particular if the true source of the cause for the failure is known. Please email me via the link on the board.
Thanks,
Phil
Do you know for a fact that the merc relay failed or was it in fact the control signal to the relay that stuck on? It is my understanding that a merc relay turns off from gravity thus cannot fail in the on position. The only way it could fail on is that the enegerizing coil was on thus the failure was actually one step if front of the relay not the relay itself.
I've had a number of side discussions with some people on this topic as I'm working on my kiln design as I speak. True, as all have said, if your kiln is placed in a safe envrionment (no flamables close by, good ventilation, etc.) the worse that can happen is that you have a puddle of glass on the floor and some damaged insulation along with a nice power bill. That being said I really don't wish that to happen. Also, I wonder as I mention above, how much "root cause" analysis is done wrt the actual failure mechanism. I reversed engineered my Paragon control electronics on my GL24. For starters the "real" schematic is nothing like what is published on the Paragon web page. That schematic is a joke. Looking over the real schematic there are 3 main failure points (and some smaller ones) that could cause one or more elements to "run away". My point is, that when a kiln goes bonkers there could have been a number of reasons for the event. Sometimes, especially if the cause is an intermittent failure of a component, it is very difficult to determine the cause as the problem is not repeatable. There is also one other cause that has not been mentioned which is plain operator error. Nothing broke, the controller did exactly what it was told to do, the operator simply told it the wrong thing. I am mildly dislexic, especially with numbers. It would be real easy for me to slip a digit here or there. I usually double check my programs after I have entered them because of this.
This is an interesting problem. Both from an analysis and feedback point I would be interested in hearing firsthand from anyone out there who has had/seen actual kiln meltdowns and in particular if the true source of the cause for the failure is known. Please email me via the link on the board.
Thanks,
Phil
Re: Root cause analysis
Phil Hoppes wrote:Brock,
Do you know for a fact that the merc relay failed or was it in fact the control signal to the relay that stuck on? It is my understanding that a merc relay turns off from gravity thus cannot fail in the on position. The only way it could fail on is that the enegerizing coil was on thus the failure was actually one step if front of the relay not the relay itself.
No, not positive yet, that seems to be the main contender.
I'll try and find out what the actual culprit was. Brock
My memory is so good, I can't remember the last time I forgot something . . .
-
Cindy next door
- Posts: 64
- Joined: Mon Mar 10, 2003 12:54 pm
- Location: Mesa, Arizona
It is sure nice to know that a fire should not start. That has always been my main concern.
I like the fact that my Skutt has a removable bottom, so that if I did have a meltdown, I wouldn't have to replace the whole kiln (at least I think so).
Like Amy, I have a lot of long firings, sometimes 24 hours and if the relays can fail going up or down, my checking when it should be on the down ramp has been a waste.
I like the point that in a 4 relay kiln one relay failure shouldn't put that element over 500 degrees. I have already had a relay fail and now sometimes I get hung up at a certain temp, so now I wonder if another is having a problem. Time to call the kiln guy.
Thanks for the detailed info!
I like the fact that my Skutt has a removable bottom, so that if I did have a meltdown, I wouldn't have to replace the whole kiln (at least I think so).
Like Amy, I have a lot of long firings, sometimes 24 hours and if the relays can fail going up or down, my checking when it should be on the down ramp has been a waste.
I like the point that in a 4 relay kiln one relay failure shouldn't put that element over 500 degrees. I have already had a relay fail and now sometimes I get hung up at a certain temp, so now I wonder if another is having a problem. Time to call the kiln guy.
Thanks for the detailed info!
Cindy next door
-
David Williams
-
Avery Anderson
- Posts: 86
- Joined: Sun Mar 09, 2003 6:21 pm
- Location: Cheshire. Oregon
- Contact:
Re: Root cause analysis
Hi Phil:Phil Hoppes wrote:Brock,
Do you know for a fact that the merc relay failed or was it in fact the control signal to the relay that stuck on? It is my understanding that a merc relay turns off from gravity thus cannot fail in the on position. The only way it could fail on is that the enegerizing coil was on thus the failure was actually one step if front of the relay not the relay itself.
Phil
This one was a puzzlement to me also. We checked the program that was entered and it was correct. The owner called Danver and they said it was a mercury relay failure. However, this made no sense to me since the controller was still operational. We had programmed it for a high of 1175...a member of the studio walked by at 10:00 PM and saw a red glow and the temp was reading 1650! Not knowing anything about kilns, he simply left it alone. The next morning when we arrived, it was down to 300. As I said...this was all very strange. There is a theory that there was an electrical spike which caused the controller to fail in the "off" position....which means it would continue to ramp up.
I'm no expert, but I've had my share of meltdowns. My ex-husband built my controller from an Omega processor. Originally, he underpowered the solid state relay and installed an inadequate finned heat sink. After three major failures...one resulting in the necessity of a new kiln lid....the SSR was upgraded as was the heat sink. Brad Shute helped us configure a shunt trip circuit breaker so I can now program my controller to shut down if the temp goes over what I program in for a high. It's still not fail safe as the thermocouple could still go, but it's sure a lot safer than it was.
Good luck with you new kiln...the beginning photos looked wonderful.
Avery
-
David Williams
Re: Root cause analysis
The owner called Danver and they said it was a mercury relay failure. However, this made no sense to me since the controller was still operational.
Think of it it this way--the controller is the brain and the relay is the muscle. The muscle might cramp up but that doesn't mean anything is wrong with the brain. I know the controllers with 110 coils do have a switching contactor inside the controller before the relay and that could fail closed theoretically like Phil said. They have a more finite life than the mercury relay itself. An engineer once told me that it is rare for mercury relays to fail on but it can happen.
I also had a meltdown because I didn't keep an ssr cool enough. I wouldn't buy an ssr again. As an aside I really love my new phase-angled scr. I have it on my furnace and I can dial the current to whatever I want. That's good because I didn't have to commit to a high amp dedicated circuit. When I'm charging I just make sure I'm not using anything else on the circuit and I dial the amperage up. When I'm idling I turn the amps down and I can run another oven on the same circuit. Its a very good system if one has a smaller service. It makes no noise except for a hum. It cost like 500$ but I made up for it with my 129$ Fuji. Which though inexpensive has a gazillion more esoteric features than any othercontroller I've seen. I still have a little skutt with an ssr but I hardly ever use it. The rest of my ovens are mercury.
Think of it it this way--the controller is the brain and the relay is the muscle. The muscle might cramp up but that doesn't mean anything is wrong with the brain. I know the controllers with 110 coils do have a switching contactor inside the controller before the relay and that could fail closed theoretically like Phil said. They have a more finite life than the mercury relay itself. An engineer once told me that it is rare for mercury relays to fail on but it can happen.
I also had a meltdown because I didn't keep an ssr cool enough. I wouldn't buy an ssr again. As an aside I really love my new phase-angled scr. I have it on my furnace and I can dial the current to whatever I want. That's good because I didn't have to commit to a high amp dedicated circuit. When I'm charging I just make sure I'm not using anything else on the circuit and I dial the amperage up. When I'm idling I turn the amps down and I can run another oven on the same circuit. Its a very good system if one has a smaller service. It makes no noise except for a hum. It cost like 500$ but I made up for it with my 129$ Fuji. Which though inexpensive has a gazillion more esoteric features than any othercontroller I've seen. I still have a little skutt with an ssr but I hardly ever use it. The rest of my ovens are mercury.
-
Kevin Midgley
- Posts: 773
- Joined: Mon Mar 10, 2003 11:36 am
- Location: Tofino, British Columbia, Canada
Just had a mechanical Allen Bradley relay wear out. Great relay with 24 volt coil. ARC WELDED ITSELF TOGETHER IN THE ON POSITION. That is it for me with mechanicals. I would use the Bradley relays again but I am not into having to preventively replace them every year or so. I am going to convert to Mercury relays for peace and quiet. I fire, watching the kilns. Didn't lose anything.
Brock only warped stainless steel, I once melted it in a pottery kiln.
You have to keep observing firings and checking that the "cat" didn't knock a thermocouple out of a kiln. Kevin
Brock only warped stainless steel, I once melted it in a pottery kiln.
You have to keep observing firings and checking that the "cat" didn't knock a thermocouple out of a kiln. Kevin
-
Phil Hoppes
- Posts: 298
- Joined: Wed Mar 12, 2003 2:20 pm
- Location: Overgaard, AZ
Avery/Brock,
If the controller was still operational and the oven was cycling down, ie you mentioned that it was at 300 in the morning, this indicates to me that the relay was in fact operating. In this case I would put my money on an intermittent thermocouple or possibly something wrong on the controller board. If the controler was getting incorrect information wrt temperature, it would keep cranking the oven up until it saw the "temp" it was looking for, all be it the wrong temperature. An intermittent failure could be quite hard to detect as what may be happening is the thermocouple could be changing its characteristics above a certain temp, say 1000 for example. Below this temp everything works fine but above lets say it starts to read 1/2 degree off for every degree increase so when the kiln is actually at 1500 degrees it is telling the controller it is at 1250 degrees. This would cause the kiln to look like it ran away, may make you think that the relay stuck and at temperatures below a certain threshold the whole thing behaves like it is fine. I'd swap thermocouples and check the kiln again before I went out and started swapping relays, especially Mercury relays. The failure does not make sense without more investigation from my standpoint.
Phil
PS - The kiln is coming along. We will be building the two bases for it probably next weekend. At the moment I'm still in construction hell. My drywall people left at 4pm yesterday with only 8' of tape left to do to complete the job. Why they could not stay 5 minutes to complete the job I will never know. I'm one who always wants to look for the best in people and think that everyone has potential but I'm beginning to think that some people really are just plain stupid and there is no cure................
If the controller was still operational and the oven was cycling down, ie you mentioned that it was at 300 in the morning, this indicates to me that the relay was in fact operating. In this case I would put my money on an intermittent thermocouple or possibly something wrong on the controller board. If the controler was getting incorrect information wrt temperature, it would keep cranking the oven up until it saw the "temp" it was looking for, all be it the wrong temperature. An intermittent failure could be quite hard to detect as what may be happening is the thermocouple could be changing its characteristics above a certain temp, say 1000 for example. Below this temp everything works fine but above lets say it starts to read 1/2 degree off for every degree increase so when the kiln is actually at 1500 degrees it is telling the controller it is at 1250 degrees. This would cause the kiln to look like it ran away, may make you think that the relay stuck and at temperatures below a certain threshold the whole thing behaves like it is fine. I'd swap thermocouples and check the kiln again before I went out and started swapping relays, especially Mercury relays. The failure does not make sense without more investigation from my standpoint.
Phil
PS - The kiln is coming along. We will be building the two bases for it probably next weekend. At the moment I'm still in construction hell. My drywall people left at 4pm yesterday with only 8' of tape left to do to complete the job. Why they could not stay 5 minutes to complete the job I will never know. I'm one who always wants to look for the best in people and think that everyone has potential but I'm beginning to think that some people really are just plain stupid and there is no cure................
well, actually you can get computers that do have redundancy. you just have to be willing to pay for that. they are getting cheaper all the time. it used to be you'd have to pay double for this; it's now down to about 1.2 times. http://www.stratus.comMarty Daily @ CDV wrote:your computer does not have redundancy by doing dual writing to 2 main disc. So why should a kiln company have to build in redundancy standard.
Now optional redundancy is a whole new issue. Custom built redundancy is availble from Jen-Ken no problem, Paragon to some extent, and Skutt to a much lesser extent, but be prepaired to pay for it. Another reason that addons are so expensive is that a kiln has to be pulled of the normal assemly line and go to a seperate person for the custom addons this becomes time consuming and drives up the cost.