DIY Composite Curing Oven

Discussion in 'Techniques' started by JLebow, Feb 14, 2019.

Help Support The Rocketry Forum by donating:

  1. Feb 14, 2019 #1

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    I've scratch build 6 fiberglass and carbon fiber rockets with parts that I've rolled and hand laid. My wife has banned the uncured epoxy from the house because of the (slight to me) smell. In the winter time, my unheated garage is cold and it takes a long time for the slow epoxy to cure. Now a few strategically placed heat lamps would probably solve my immediate problem. But, eventually I would like to get into some of the composites that stand up to the heat that Mach 2+ flight can produce. Those higher temp epoxies typically specify a one or two step curing profile between 120 and 250 degree F.

    So this is what I built.

    I started out wanting a metal liner. After looking at raw material prices, I found this locker for about $80. Mine was grey. It was the exact dimensions I had on my sketch pad: 6 feet x 18 in x 12 in.
    locker.jpg
    The first job was to plug up all of the gaps: holes punched along the side walls for adjustable shelves, ventilation louvers, panel joints, and areas were the legs exit the locker body plugging all the holes.jpg

    100 pop rivets and a tube of high temperature RTV gasket maker later, I had the making of a fairly decent oven liner.

    Industrial heating elements are expensive. Single burner hot plate stoves are not. I stripped out all of the plastic parts, the thermostat, and indicator lamp, and wired pigtails to the heater element with heat rated ring terminals and high temperature wire.
    heater element.jpg
    I don't want to burn down my house, so I also included a thermal cutout fuse that opens at 450 degrees F, wired in series with the heater element. It is mounted in a ceramic terminal block inside the oven.
     
    prfesser likes this.
  2. Feb 14, 2019 #2

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    I decided that I wanted a blower to provide some kind of air circulation. With such a long enclosure, I'm sure there would be a significant temperature gradient. I probably should have gotten a convection fan from a commercial oven. This would have been rated for max 300 degrees F I expect out of this oven. Instead I got on of these blowers. stock blower.png

    It supposedly can push 12 cfm with zero static pressure. But I don't need a lot of flow. To deal with isolating the motor from high heat I ordered these parts: blower parts.jpg

    We are going to make this thing chain driven. The new blower fan shaft is riding in a pair of ball bearings. It turned out like this: blower in profile.jpg blower completed.jpg

    I made the bearing end caps and collars that limit shaft end play with my lathe. For ducting, I have 5 feet of flexible 1.5" diameter exhaust pipe. I make the flanges on both sides of the pipe and sealed them to the locker with high temp RTV.
    blower mounted.jpg blower system.jpg
    The flange for the copper elbow is still a slab of aluminum rod in this picture. It was bored on the lathe, and bolted to the locker to seal up the copper elbow.
     
    Last edited: Feb 14, 2019
    prfesser likes this.
  3. Feb 14, 2019 #3

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    Reserved for controller
     
  4. Feb 14, 2019 #4

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    Reserved for final construction
     
  5. Feb 14, 2019 #5

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    I guess I outsmarted myself here a bit. When I returned from putting the kids to bed, the edit button seems to have disappeared (timed out?).

    The controller is based on an arduino uno microcontroller (adafruit version), a 3.5 inch touch screen, and initially a thermocoupler (later switched to an RTD). Here it is bread boarded, after I first got the temperature reading function to work, and the LCD initialized. controller on breadboard.jpg

    So I wanted to be able to program several temperature profiles and store them in the non-volatile memory on the microcontroller board. These are entered as several target temperature and duration pairs.
    Example:
    100 F, 20 minutes
    100 F, 60 minutes
    150 F, 20 minutes
    150 F, 75 minutes

    The controller will slowly ramp the target temperature to 100 degrees from the current temperature over a period of 20 minutes, then hold the temperature for 60 minutes. The the temperature target will ramp to 150 degrees, over a period of 20 minutes, and hold 150 degrees for 75 minutes.

    The microcontroller then compares the measured temperature with the target temperature and uses a function I coded called PID control to compute how long the heater element should be turned on and off (duty cycle). The microcontroller is connected to a 25 amp solid state relay, that is wired in series with my heater element. Here are the guts of the controller box (controller, solid state relay, fuse, big red start button):
    controller in box.jpg

    I also wanted the display to have a graph of the setpoint profile and overlay the measured temperature.
    controller ramp soak.jpg
    This is the result of a two step ramp to 140 and 170 degrees F after a couple cycles of tuning my PID control loop. You can see a bit of delay lag at the start of the heat cycle. It take 4 minutes of cycling the heater element before the temperature sees any increase in temperature (thank you thermal mass). This time lag is the enemy of a good control loop. I'm actually OK with the 3 degrees over overshoot that occur at the two plateaus. This is an overshoot of the air temperature in the oven, and my composite parts will have a bit of their own thermal mass, so this amount of overshoot is going to be damped in my part, and may actually speed up their equilibrium during the soak.
     
    prfesser likes this.
  6. Feb 14, 2019 #6

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    I decided to insulate with mineral wool. The material is rated for 2000 degrees F, and wasn't too expensive. The material is 3 inches thick and should not be compressed to work well as an insulation. I framed a box with 1x2 inch furring strips, and clad it with 3/8" thick plywood. insulated box.jpg
    insulated box 2.jpg insulated box 3.jpg

    Earlier I mentioned that I did not stick with the thermocouple. I was seeing about 15 degrees of measurement error that appears to be due to electrical noise from the blower fan motor. A K type thermocouple produces a 80 uV per degree signal, so it it not hard to imagine noise causing a problem. I switched to a platinum based resistive temperature device (PT1000 RTD). This device has 1000 ohms of impedance at 0 degrees C and the resistance goes up as the platinum element is heated. The change is much more immune to the electrical noise in my setup.

    I guess the next step is to build some more rockets. I invested a good portion of this years build season to making this oven, and the first launch of the year is only 2 months away.
     
    prfesser likes this.
  7. Feb 14, 2019 #7

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    And clean up the shop. As you can see in the picture, that place is wrecked.
     
    prfesser likes this.
  8. Feb 14, 2019 #8

    AeroAggie

    AeroAggie

    AeroAggie

    Forum Supporter TRF Supporter

    Joined:
    Jan 21, 2017
    Messages:
    200
    Likes Received:
    32
    Gender:
    Male
    Location:
    Wichita, KS
    Nice work!

    Do you vacuum bag your parts? Might consider plumbing the box with internal vacuum lines and a bulkhead fitting.
     
  9. Feb 14, 2019 #9

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    Hey thanks. Yeah vacuum is part of the plan. Here is what I've put together so far. Pump, buffer tank, check valve, vacuum switch adjustable between 5 and 14 psi, and power relay. I found this site to be really helpful for finding the odd part. I don't have an interface to the oven worked out yet. vacuum.jpg
     
  10. Feb 16, 2019 #10

    G_T

    G_T

    G_T

    Well-Known Member

    Joined:
    Dec 29, 2011
    Messages:
    2,017
    Likes Received:
    116
    Fairly fancy!

    We used to just use foam insulation board from the hardware store, and incandescent light bulbs or a small oil circulation heater as the heat source. I think small ceramic heaters have been used as well. IIRC, we used a thermometer and manually adjusted for temp.

    Of course what we did would not have been suitable for heat curing / post curing Cotronics resin or some other high temp options!

    We were using MGS 285 epoxy (which is about as high temp as the non-high-temp systems go) so didn't need to get it all that hot for a good final Tg. 122F service temp without heat treat, 176F with heat treat. http://www.ezentrumbilder.de/rg/pdf/td_en_LR285_eng_180310.pdf It's wonderful laminating resin to work with. The technical data available for aerospace epoxy systems helps with determining how you want to process. IMHO, if the data isn't available, use something else!

    I liked having a manifold between the vacuum pump and the bags. https://www.rocketryforum.com/attachments/gopr0331-fixed-jpg.106416/ (From my Sprite and a baby O thread). Included in there is a filter to help remove volatiles (water, other solvents). You want some sort of filter between the epoxy end and the pump to protect the pump over time. The gauge lets you know the vacuum you are drawing at the manifold but realize the vacuum downstream will not be quite as high. The valves let you run several parts simultaneously, and independently, which is quite useful when you want to do more than one part at a time. This sort of thing was pretty common.

    Gerald
     
  11. Feb 17, 2019 #11

    sooner.boomer

    sooner.boomer

    sooner.boomer

    Well-Known Member

    Joined:
    Mar 21, 2011
    Messages:
    2,401
    Likes Received:
    74
    One thing I really worry about in oven projects is safety. I've designed/built a lot of controller systems, and it never hurts to be paranoid (but, am I paranoid *enough*?). What I would add are; fire/smoke alarms, a nearby fire extinguisher, and a thermal safety fuse. It wouldn't hurt to have a second temp display for safety, too. This doesn't have to break the bank. Here are some parts and vendors that are great for simple DIY stuff:

    BG Micro bgmicro.com

    BG Part Number: FUS1132 $0.65
    Non resettable, axial, upper-limit temperature protection thermal fuses. Over 10 Amps continuous current. 228 degrees C. (442.4 F)

    BG Part Number: TST1201 $5.95
    Digital Pyrometer 0-1999º F ( 0-1200º C ) K Type Input
    This pyrometer runs on a 9V battery and measures high temperatures in both Fahrenheit and Celsius. On/Off switch attached by leads, switch for Fahrenheit and Celsius mounted on back.

    All Electronics allelectronics.com

    CAT # THM-4 $6.00
    THERMOSTATIC CONTROL MODULE
    Digital module can be programmed to turn equipment on or off at specific, pre-set temperatures. Also has hysteresis setting, delay setting, upper-limit setting and upper limit alarm setting. Displays temperature in Celsius on 3-digit LED display. On-board 20A relay. Waterproof temperature sensor. Three buttons for navigating temperature parameters and other functions. Operating voltage 12 Vdc.
     
  12. Feb 17, 2019 #12

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    Solid state relays do tend to fail closed, so I don't think being extra careful is a bad idea. I did a bit of thinking on the potential failure modes, but didn't share much about that in this thread.

    I bought a solid state relay from a brand I recognized (Omega), that had zero crossing sensing for switching the load, and selected an over-sized heat sink to keep it from overheating.

    I do have the the smoke detector, fire extinguisher, and thermal fuse (mine cuts out at 216 C: 3 PCS New SF214E NEC SEFUSE SF214E Thermal Fuse Thermal Cut Out 216C Degree, 10A 250V) already in place.

    I also verified my controller temp readings with a handheld thermocouple meter. It measured to within a few degrees.

    The commercial thermostat rigged to cut out in overtemp mode is a good suggestion. I will see if there is something like your linked item, but good up to 300 degree F. Thanks for voicing your concerns.
     
  13. Feb 17, 2019 #13

    sooner.boomer

    sooner.boomer

    sooner.boomer

    Well-Known Member

    Joined:
    Mar 21, 2011
    Messages:
    2,401
    Likes Received:
    74
    Belt AND suspenders...
     
  14. Feb 20, 2019 #14

    JLebow

    JLebow

    JLebow

    Active Member

    Joined:
    Oct 31, 2017
    Messages:
    25
    Likes Received:
    5
    Looks interesting. I'll give it another look when I'm ready to order more epoxy.
     

Share This Page