DIY Heat Treat Oven Build

[Cushing H.]

putting those steel rods through the top bricks seems a good idea! I was wondering how one might span multiple bricks across the top....

3 kW seems a lot for that chamber size - it must come up to temp (and recover) pretty fast!

 

[Luis Barros]

putting those steel rods through the top bricks seems a good idea! I was wondering how one might span multiple bricks across the top....

3 kW seems a lot for that chamber size - it must come up to temp (and recover) pretty fast!

Takes 5 minutes to reach 1000F then takes another 20 minutes to reach 1550 and 1 hour and 30 minutes to reach 1950F.

edit:temperature times, yesterday I decided to be exact with the temp/time, fast to reach 1000F but then takes more than an hour to reach 1950F

 

[scott kozub]

25 soft bricks
kanthal A1 16 gauge (59'), 3kw, 240V.
Inside dimensions W=8.5 L=15.5 H=6.75
PID - Inkbird ITC-106VH with 2 SSR's
3M oven sealant (2000F) - the top bricks I also drilled steel rods between the bricks in case the sealant fails.
The bricks was the most expensive item which cost me $120 for 25 bricks everything else was not bad with the PID and the wire $30 each and the sealant was $15 for 2 tubes.

Keep an eye on those coils. When they heat up for an extended period they will sag and short out. Once they touch they have a bit of a tendency to weld together and are very hard to get back into place when cold because they're brittle. Guess how I know.

The build looks very good though.

 

[Natlek]

25 soft bricks
kanthal A1 16 gauge (59'), 3kw, 240V.
Inside dimensions W=8.5 L=15.5 H=6.75
PID - Inkbird ITC-106VH with 2 SSR's
3M oven sealant (2000F) - the top bricks I also drilled steel rods between the bricks in case the sealant fails.
The bricks was the most expensive item which cost me $120 for 25 bricks everything else was not bad with the PID and the wire $30 each and the sealant was $15 for 2 tubes.

Make and use some staples to hold elements on place ....Don t make them in shape of U , make them in shape of J .....with one leg short and other much longer ,they hold better.........

 

[JTknives]

What type/brand are those bricks, Thy look like pumice. All the Morgan k23 bricks I have used look different then those.

 

[Luis Barros]

Make and use some staples to hold elements on place ....Don t make in shape of U , make them in shape of J .....with one leg short and other much longer ,they hold better.........

That's a good idea, I installed every 2 inches a staple with a V shape but I can see some moved a bit.

 

[Luis Barros]

What type/brand are those bricks, Thy look like pumice. All the Morgan k23 bricks I have used look different then those.

I bought them from a pottery store in Pennsylvania and they have listed as soft brick, IFB 2300, 2.5" and they sale them individually or by a box of 25 ($119.50)

 

[Luis Barros]

Keep an eye on those coils. When they heat up for an extended period they will sag and short out. Once they touch they have a bit of a tendency to weld together and are very hard to get back into place when cold because they're brittle. Guess how I know.

The build looks very good though.

I'm going to put more staples just in case.

 

[Luis Barros]

When I tried to keep the temperature at 200F I see huge fluctuations, the element goes off (I have a light showing when the element is on/off) at 200F but the temperature keeps going up almost to 300F then starts going down and even the element goes on still goes down to around 170F, is this a problem with the PID settings?

 

[Cushing H.]

When I tried to keep the temperature at 200F I see huge fluctuations, the element goes off (I have a light showing when the element is on/off) at 200F but the temperature keeps going up almost to 300F then starts going down and even the element goes on still goes down to around 170F, is this a problem with the PID settings?

The pid settings for operating at 200F will be very different from those that work at 2000F. I presume that you have it tuned for 2000F?

bottom line: with your huge power input, at 200F you will need a really wide proportional band. How you set that depends on the controller

 

[Luis Barros]

The pid settings for operating at 200F will be very different from those that work at 2000F. I presume that you have it tuned for 2000F?

bottom line: with your huge power input, at 200F you will need a really wide proportional band. How you set that depends on the controller

So that's the problem I AT with however the temperature was set which probably was 60 or 70 when I installed....Warning user error I will try tonight with the temp set at 1850F ( I have D2 steel I want to try).

 

[JTknives]

So that's the problem I AT with however the temperature was set which probably was 60 or 70 when I installed....Warning user error I will try tonight with the temp set at 1850F ( I have D2 steel I want to try).

Make sure to wrap your D2 in stainless foil before you toss it in the oven.

 

[scott kozub]

I think what was said above is right. Your design sounds similar to mine and it heats up very fast but mine doesn't run well at low temps. I have a toaster oven for tempering.

Remember that the coils are just on and off like a light switch. When they're on there s a big heat input and the coils get way hotter than the desired temp. At higher temps this is much less of an issue.

 

[JTknives]

I think what was said above is right. Your design sounds similar to mine and it heats up very fast but mine doesn't run well at low temps. I have a toaster oven for tempering.

Remember that the coils are just on and off like a light switch. When they're on there s a big heat input and the coils get way hotter than the desired temp. At higher temps this is much less of an issue.

this is why my new oven is being proportionally controlled and not standard pid controlled. I have one big tr
Peeing oven that uses this system and it’s amazing.

 

[Cushing H.]

There is something here that is not making sense to me. If you are using a PID controller, AND you have an SSR feeding electricity to the elements, AND you have truly set up the PID with a suitable proportional bandwidth (so that the element is quickly cycled on and off in fractions of a second, with fractional percentage of "off" time increasing as you approach your set point) .... then you to all intents and purposes have a true proportional control on that oven. This makes me wonder if the PID is being configured correctly????

 

[JTknives]

There is something here that is not making sense to me. If you are using a PID controller, AND you have an SSR feeding electricity to the elements, AND you have truly set up the PID with a suitable proportional bandwidth (so that the element is quickly cycled on and off in fractions of a second, with fractional percentage of "off" time increasing as you approach your set point) .... then you to all intents and purposes have a true proportional control on that oven. This makes me wonder if the PID is being configured correctly????

From my experience this is about half true. When a true proportional control auto tunes it know what levels for what temps. It’s cycle time is very much like a dimmer switch so it’s using PWM. Pulse width modulation is cutting off the cycle x% into the ac cycle. So it’s running at 60hz. Where a normal pid is cycling quite fast when it’s cycling 1-2 times a sec. Some pids work better then others which has to do with how thy have set up their algorithms. This is what bugs me about the TAP controller, it does not have an auto tune. So you left fiddling with the setting in the advanced menus (which require a password ). The pid controllers from Auburn’s are quite nice for the price. I have kinda switched over to watlow for my temp co trolling as I really love their stuff. Also their tech support is AMAZING. Last time I called it rang once and was treated with a “hello”. Took me by surprise and I had to confirm it was watlow I called. He spent 20min with me sorting out a 480 to 240 conversion on an oven I bought second hand. Thy did not build the oven it just had their control components in it. That sold me as it’s rare to get that kind of service anymore.

 

[Cushing H.]

From my experience this is about half true.

I get it. My experience has been with commercial-level PID's which are able to cycle quite fast in the proportional band. If the inexpensive ones typically used by Makers are restricted to longer cycle times, then might explain some problems. However - the one controller I have had a chance to study the manual for in-depth (Auber SYL-2362A2) does clearly indicate you can control the cycle time (parameter is "ot") down to quite short cycle times much much less than a second. You need to go manual in configuring the parameters, though, to do that. I wonder if Autotune is getting in the way of getting better results?

This leaves me wondering, again, if it is worth my putting together a thread on PID's and how to tune them. People seem to so like the autotune capability though that I still do not know if that would be just wasted electrons...... (a joke - electrons wasted in storing the data.......)

 

[scott kozub]

I get it. My experience has been with commercial-level PID's which are able to cycle quite fast in the proportional band. If the inexpensive ones typically used by Makers are restricted to longer cycle times, then might explain some problems. However - the one controller I have had a chance to study the manual for in-depth (Auber SYL-2362A2) does clearly indicate you can control the cycle time (parameter is "ot") down to quite short cycle times much much less than a second. You need to go manual in configuring the parameters, though, to do that. I wonder if Autotune is getting in the way of getting better results?

This leaves me wondering, again, if it is worth my putting together a thread on PID's and how to tune them. People seem to so like the autotune capability though that I still do not know if that would be just wasted electrons...... (a joke - electrons wasted in storing the data.......)

To your point above your probably right. I have never taken the time to play with the pid because it was working well at the high end. I would love a good write up on pid setup as the instructions that came with the ink bird are garbage.

 

[scott kozub]

The large swings in temperature at the low end could also be due to the inexpensive sleeves thermocouple with a slow reaction time?

 

[Cushing H.]

The large swings in temperature at the low end could also be due to the inexpensive sleeves thermocouple with a slow reaction time?

Actually i would think not. The physics of the thing is based on temperature differences between the thermocouple and its environment, not absolute temperature. (So something like 410 outside versus 400 inside (inside the thermocouple sleeve) would react just as fast as say 2010 F outside versus 2000 inside).

what, to my ear, often gets missed here, is that a pid can not be considered as “enough” just by itself. In reality the pid and the thing it controls are a system ... what you do to the device affects how the pid reacts and vice versa. (People here have bumped up against that when they got better results with insulating a toaster oven, and we are seeing it again here with the monster heating elements in this oven build). Autotune can help a little - but if things are too far out of balance autotune will not work well. But i am at risk of hijacking this thread - so will stop here...