Sciencemadness Discussion Board - Electrical Furnace Contruction - My design and implementation

Electrical Furnace Contruction - My design and implementation

Pages: 1 2

12AX7 - 27-3-2006 at 10:29

Well lemme see here. 10" dia., 10" tall, say 15 turns, wire gauge doesn't matter but call it #20 (AWG for Cu I think), stainless steel resistivity (close enough), the calculator program I have shows 44uH and 5.2 ohms DCR (not unreasonable). At 60Hz, that's 0.017 ohms reactance, or a Q of 1/313th. (An intentional inductor has Q = 50 or better, a difference of 5 orders of magnitude or so! :cool:

)

One thing you'll notice is, for a solenoid coil, you will have 5 or 10 or 50 amps through 10 or 20 turns, and that's 50 to 1000 amp-turns, enough to make a piece of iron buzz when energized.

(My induction heater is currently around 1kW power output and runs up to 2000 amp-turns at 21kHz, enough to make a piece of aluminum or copper, placed in the coil, stand on end from the repulsive force!)

Tim

Cloner - 27-3-2006 at 12:20

OK, nothing beats an example

Thank you.

Magpie - 19-9-2006 at 11:14

I have been wanting to build a tube furnace for some time. This would be a significant project using a horizontal clamshell design with castable insulating mortar and a castable refractory. It would also have a temperature controller and be capable of developing 1300C.

I decided to build a small scale furnace as a first step to get some experience with hightemperature materials and fabrication techniques, as well as provide a furnace for another project. I built this furnace as a "vertical tube" furnace using a 3lb coffee can as shell. The dimensions are: H=7.5", OD=6", and ID=2.5". The refractory was made using axehandle's formula. The heating coil is a 5000w, 240VAC, D = 0.5", L =6', salvaged from one of the banks of the furnace that formerly heated my house. For my coffee can furnace I run it at 120VAC, 1250w.

When power is applied it comes up to a red-orange heat. Since the insulation is not very thick the shell of the furnace gets pretty hot also. This is not good and will likely limit its lifespan. I have not kept it at full power for more than a few minutes so far. I have attached a picture below, and one in the next post.

I also am building a temperature controller for this furnace which will be suitable for any furnace revisions as well as the horizontal tube, clamshell type that I want to eventually build. The controller framework is shown along with the top view of the furnace in the next picture. It is complete except for two components that should arrive any day now: (1) a solid state relay, and (2) a PID controller with type K thermocouple. (These controllers now are surprisingly cheap.)

A few comments on axehandle's refractory formula: Making/using this was different from any of my past mortar, grout, or concrete experience. First of all it was very fluid. This is contrary to my other experience where I only used enough water to make the material workable. Second, the pearlite floated to the top and the sand sunk to the bottom so the mix was not homogeneous. So I just let it set for a few hours, occaisionally stirring and checking its stiffness. Eventually it did stiffen enough that I felt it was reasonably homogeneous but still pourable and workable. After hardening it is not strong like concrete. But so far it has only developed minor cracks and seems to be doing the job for which it is intended. Until someone else comes up with a better formula this is the best we have, and is cheap, OTC, and at least so far, is doing the job for me. So .....thank you, axehandle.

Magpie - 19-9-2006 at 11:19

Here's the 2nd picture:

furnace + controller.jpg - 175kB

Magpie - 26-9-2006 at 19:34

I just received the PID (with auto-tuning) controller for my furnace temperature controller. I connected everything up and tried it out by controlling the temperature of the outside of a lamp bulb to 100F. It works well. This controller should be suitable for most any furnace. I assembled this controller for about $100.

[Edited on 27-9-2006 by Magpie]

temperature controller.jpg - 218kB

12AX7 - 26-9-2006 at 22:28

Those Watlows are nice, the one that's four times bigger (92 or 93?) are awesome, programmable, self-adjusting, the works. Can't say I'd like to pay for one though!

Tim

Al Koholic - 27-9-2006 at 18:55

So guys, I'm also building a furnace with a lot of help from all the great work that has been done on this thread. Thanks all for the contributions!

I've got most of it hashed out and I have almost all the parts now but there is something I think we need to clear up.

How are you all connecting the element wire to the power leads? Not like series or parallel...but what kind of connectors are you using?

I ESPECIALLY like Magpie's ceramic(?) connectors. Magpie...I fear you got those from the salvaged furnace. If not, please tell me where you got them! Many google searches have yielded little in the way of solutions. Definetly nothing as elegant as what you have there.

Magpie - 27-9-2006 at 19:41

Al, those ceramic rings did come from my furnace. They really are just sort of there for looks in my little furnace. A bolt runs through them and the wires are just secured with washers and the bolt, and a nut. The element ends transition from the dark, rough looking coil wire to a clean shiny metal wire end that looks like it might be made of nickel. This is for about 1" on each end. That's the way I found the coil.

When I build my high temperature furnace I will likely be using 1mm diameter Kanthal A-1 to form my heating coils. I will likely just have to lead the wire ends outside the furnace wall a couple inches for terminals. I don't know what else to do. This is what rikkitikitavi recommends early on in this thread. Hope this helps.

Had a feeling...

Al Koholic - 28-9-2006 at 20:45

Well I had a feeling they were from the furnace. I was at Home Depot today and someone mentioned a few stores I should check out so I'll try and find those.

It wasn't so much I was wondering how physically to do it, but more what the actual connectors you all were using. I guess I'm just in love with the way those look on your furnace Magpie.

I've got the regulator box with fans basically ready to go soon and just have to get my element wire in the mail. I'm using K-A1 20AWG (.8mm). Should be pretty nice, I'm pumped. Too many plans to use the thing for.

Knife making (hopefully pattern-welded "damascus" style can be acheived with it) is one of the main reasons I am building the thing in the first place. No need for worrying about flame quality or fuel quantity anymore!

I suppose I'll have to figure out how to upload some pics of my completed setup when all is said and done.

Magpie - 29-9-2006 at 08:48

I've seen little ceramic parts (tubes, etc) on some websites for sale but can't remember for sure which ones. I think I looked under "ceramic insulators" and found some such. They are available, just difficult to find as they are so specialized. Pottery and kiln shops would also be worth a look.

[Edited on 29-9-2006 by Magpie]

tumadre - 29-9-2006 at 08:59

For knife making you still want a reducing gas mix in side the furnace

i have used 17 gauge electric fence wire (15 gauge AWG), and for some smaller stuff I have used 1/8 inch steel rod as an alternative to nicrome, it don't last but 10-40 hours, but the cost is nothing.

as the bolt goes, i have not tried stainless, i'm sure it or something else is needed

and don't even think about exposing copper to more than 500C

12AX7 - 29-9-2006 at 12:15

I'm interested to know how you intend to get hot enough to forge weld, without induction (hello) or fire.

Tim

Al Koholic - 30-9-2006 at 10:27

Well, steel welds at a lower temp than iron; the higher % C, the lower the welding temp. I've read that in some alloys, the welding temp can be as much as 1000F below the melting point. Usually it's around 300-400 below.

Anyway, the furnace may never reach that T, but it will still be useful for forging. All that is a long way off anyhow...

As far as a reducing gas mix, would not keeping the lid on during the heating cause the O2 to be used up by light surface oxidation forming some scale? This is typical and needs to be brushed off in any forging of course. I'm hoping that I wouldn't need to add anything extra to the furnace to eat O2 but it wouldn't be a big deal...little OT thought.

tumadre - 30-9-2006 at 12:15

All the websites I've read use a carburizing furnace because the gas reduces the iron oxide as the weld is taking place, no amount of flux will work without at least a neutral gas mix.

no amount of iron will pull the O2 out, charcoal might but i would'nt even let Nitrogen get in without a more active gas like propane floating in the furcace

if you hold the electric element internal to the kiln at 1350C, you will be able to forge the weld, but it will take too much time to re-heat the knife to 1100 or 1250C

IIRC iron melts at 1536C

aren't you pulling the knife out and beating it with a hammer like 300 or more times?

Quince - 1-10-2006 at 01:14

Quote:

I just received the PID (with auto-tuning) controller for my furnace temperature controller....I assembled this controller for about $100.

What the hell?! You can build an analog PID for $10 in parts. Plus, autotuning can never be as good as good manual tuning, and, of course, real men don't use autotuning -.-

Al Koholic - 1-10-2006 at 05:41

Tumadre, I'll have to see how it goes with the heating times. I didn't think it would be too long to do reheating but I'll have to see. Like I said, this might not be a furnace I can use for forge-welding. Out of curiosity, why would N2 be a problem in the furnace?

On an interesting side note though, no you don't have to composition to begin with. You can make a stack of say, 5 layers to start and weld them. Then you cut the result in half and restack to get to 10 layers. Then to 20, and so on. Makes getting layer counts in the 50-250 range (the range that fold the metal over all that much if you cheat by using two steels with slightly different looks good) easier than starting with a single bar and folding it.

Magpie - 1-10-2006 at 08:50

Quote:

What the hell?! You can build an analog PID for $10 in parts. Plus, autotuning can never be as good as good manual tuning, and, of course, real men don't use autotuning -.-

That PID cost $35. I'd be pretty impressed if you could duplicate that for $10 alright. It accepts all common thermocouples and thermistors with no work on my part. In my younger years I might agree with you about auto-tuning. Right now I'm very glad to have it.

bio2 - 1-10-2006 at 09:03

You can salvage resistance wire ceramic holders from a clothes dryer,

The ends are usually welded to stainless steel terminals.

Damn, I must be getting old as I don't remember PID controllers less than about $200. Been awhile but autotuning never did live up to it,s claim. Programming A manual sequence is the best way in my experience.

[Edited on 1-10-2006 by bio2]

not_important - 1-10-2006 at 17:48

Quote:

Originally posted by Al Koholic
... Out of curiosity, why would N2 be a problem in the furnace?

....

Most of the resistance wire for heating is designed to work in an oxidising atmosphere. Running them in reduction may greatly shorten their life. Running them in a neutral atmosphere, after breaking the wire in under oxidising conditions, may be OK; but if traces of reducing compounds got in you could be back to the short lifespan probem.

Maya - 7-10-2006 at 04:45

Has anybody tried or thought about making or done an argon arc furnace?

This would be for melting oxygen sensitive metals over 1700 celcius

Another idea is an IR image furnace for the same application , this focuses with mirrors, concentrated light on the sample in a quartz tube to melt it.

Quince - 7-10-2006 at 04:47

This reminds me of the infrared soldering such as http://www.aoyue.de/images/prod/Aoyue_710_1.jpg
Halogen light-type infrared source in a focusing reflective enclosure.

tumadre - 21-10-2006 at 00:32

N2 may not be a problem.

To forge a weld, all the impurities between the steels must properly flow out as the steel joins
Flux allows that to happen, but the flux can't cover all the steel at all times, so on your first few knives you manage to make, you may lose 1/3 of the steel to oxide. so you need something to displace the oxygen, and nitrogen won't do that unless you have a tank of it.

I remember somewhere someone mentioned a salt bath to heat the steel, but that may have been for heat tempering at lower temps.

Fleaker - 22-10-2006 at 09:05

Quote:

Originally posted by tumadre
I remember somewhere someone mentioned a salt bath to heat the steel, but that may have been for heat tempering at lower temps.

It would have to be at lower temperatures if you're talking about NaCl as it would have too much volatility at those temperatures. I've melted NaCl/CaCl2 before to use as an aluminum flux and it has significant volatility at only 700*C.

fuse123 - 27-10-2006 at 11:29

what about inductance furnace
any one here have any information to make it

12AX7 - 27-10-2006 at 16:19

Quote:

Originally posted by fuse123
what about inductance furnace
any one here have any information to make it

WTF, you just posted in the thread with links to many details, in fact one of approximately three such highly detailed websites on the subject on the whole internet as far as I can tell!! Did you post this before discovering the thread or are you truely that daft?

Xenoid - 3-12-2007 at 17:32

Sorry to dredge up an old thread....

I was at a recycling centre today (God, I love these places) and I picked up an old electric radiator (heater). What I was especially atracted to were the 3 identical heating elements. They are the silky silica tube type with coiled resistance wire down the centre and are about 45 cm. long. The heater doesn't look like it's had much use so they are in good condition. The heater was rated at 2400 watts so they are each 800 watts. I'm thinking of cutting each one in half or thirds, leaving the wire intact, and using them around the inside of a small muffle furnace/kiln, hopefully able to reach about 1000 oC.

My current thinking is something like a small microwave oven shell, lined with firebrick or sillimanite slabs, with the shortened elements fitted into holes and lining the interior.

Early days on this, I'll let it simmer for a while before doing anything rash, but any ideas on utilisation would be appreciated!

Image shows the 3 elements, heavy duty 3 way switch, silicone covered wires, etc. after dismantling.

Regards, Xenoid

Heater-elements.jpg - 20kB

497 - 5-12-2007 at 13:42

so i had an idea about making an induction furnace (this thread seems like the most relevent): use a electronic fluorescent ballast. would it work? i know they make 30-50 kHz which should work pretty good. i like the idea because it looks like a pain in the ass to engineer and put together the tank circuit and all that other jazz needed to build one from scratch... obviously a ballast isn't meant to power a big inductor so maybe doing so would fry it, that's waht i really need to know.

Twospoons - 5-12-2007 at 14:42

You'd do better to modify a bench-top induction cooker - you know, the portable sort you plug into the wall, and takes one pot. They can be had very cheap second hand, and they're usually good for a kW or two. You'd be lucky to find a fluoro ballast that will do more than 50W.

497 - 5-12-2007 at 15:48

ugh.. the problem with that is those cookers are very hard to get if not impossible where i am.. if i could i would love to use one. and actually i might be able to get a couple ballasts that drive two 8 foot tubes each. still not too powerful though. i'm not sure what other alternatives there are.

12AX7 - 5-12-2007 at 17:06

You're welcome to try, say, hooking up a ballast to some coils.

But you'll go through a lot of ballast units and fuses doing it.

Tim

old post, new question =]

quest - 14-9-2011 at 07:23

Hi all, I want to build an electric furnace and I have some questions. I guess it's better to use this 5 years old thread rather than open a new one.

the furnace I want to build is a 12 c"m ID furnace, mostly it will be used to melt 1 kilo batch of magnalium in a batch using tin can as a vessel for the melted metal.

I have few questions:
1) Most of the people in here used 1m"m Kanthal A-1 wire, I can get this wire in almost any size. In my furnace should I use the 1 m"m wire or a wider/thinner wire will be better?

2) For a "power supply" I wan't to use ordinary dimer used as a switch for light bulb in my house. on the dimer it said "800W" and when I opened it I saw a fuse of 4 Ampere in it. The problem is I'm not sure how long my wire should be for me to use this dimer with out ruining the 4A fuse, but still use its max
capacity (800W)?
how I tried to calculate the wire length:
P = (V^2)/2R = (I^2*R)/2
V=240V, I=4A (max A before fuse burn)
R=V/I=240/4=60 ohm
1m"m kanthal wire have a resistance of 1.76 ohm/meter.
X(wire length) = 60/1.76 = 34 meters(!)
sound too big for my small 12c"m ID furnace

, and from my understanding of electricity any shorter wire will have have less resistance ,hence more current, and will burn my 4A fuse.

3) will a 800W be enough to make an amateur electric furnace for melting 1 kilo magnalium? will it be enough to get to 1000 degrees Celsius and melt copper? glass? (just for the science of it)

thanks,
matan.

Magpie - 14-9-2011 at 08:46

Also see:

http://www.sciencemadness.org/talk/viewthread.php?tid=9705&a...

watson.fawkes - 15-9-2011 at 06:03

Quote: Originally posted by quest

Most of the people in here used 1m"m Kanthal A-1 wire, I can get this wire in almost any size. In my furnace should I use the 1 m"m wire or a wider/thinner wire will be better?

There's a reason why resistance wire is available in a really huge number of sizes, including (I've seen) half-gauge increments. The basic rule is that you take your supply voltage as fixed and your power (wattage) requirements as a given. That gives you a total resistance you need. So for each gauge of wire, you can compute the length you'll need.

There are practical limits on the specific surface power (W/m^2) that resistance wire can withstand without failing. Get resistance wire white hot in ordinary atmosphere, it will burn out. Don't do that. The effect of this is to put a minimum diameter requirement on your wire. Using the very smallest diameter wire reduces direct material costs, but also boost specific surface power and reduces service life. Depending on your application, you may want to derate your wire a couple of gauge sizes to compensate.

quest - 15-9-2011 at 10:14

thanks. now I understand it better.

one more question - how do you connect kanthal wire if it brake or if you want to connect number of kanthal coils togather? the wire will get to about 1000 degree celsius, so I can't just weld it.

Magpie - 15-9-2011 at 10:37

Quote: Originally posted by quest

how do you connect kanthal wire if it brake or if you want to connect number of kanthal coils togather? the wire will get to about 1000 degree celsius, so I can't just weld it.

Use a stainless steel crimp, available at hardware stores.

[Edited on 15-9-2011 by Magpie]

watson.fawkes - 15-9-2011 at 12:57

Quote: Originally posted by quest

how do you connect kanthal wire if it brake or if you want to connect number of kanthal coils togather? the wire will get to about 1000 degree celsius, so I can't just weld it.

I agree with Magpie. Use a crimp.

But you can weld it if you'd rather. Welding temperature is hotter than 1000 °C for Kanthal alloys. Rather than explain it myself, here's a link to a document by a guy who's done it: http://rick.sparber.org/Articles/we.pdf.

Endo - 20-1-2012 at 14:17

Using firebrick and a tile saw I was able to cut some channels to hold a nichrome element salvaged from an element for an electric furnace. I also used the insulators left from the element. The dimensions inside are 6 1/2 inches deep, 4 1/2 inches tall and 4 1/2 inches wide. I picked up a tube of refractory cement for 5$ (good to 1100C, comes in a caulking tube) and assembled all but the top brick in back. I notched the top back brick to take the insulators and decided to leave it unmoartared in case I have to change the wire out. I intend to wrap the whole brick center in a layer of fiberglass insulation (cheap) and then bend a sheet metal case for it. So far the door will just be two firebricks moartared together. It has gone together pretty easy. I hooked it up to 120V and it pulls 16 amps through an Ammeter I didn't let it get really hot because the cement wasn't completly cured yet, so it may drift a bit as it heats up. This puts the output at 1920W. I need some way to control the temp... Will a light dimmer work or will it just burn up? Back in this thread I saw a diagram for a simple control circuit but it was 220V. I also have a switch from an electric range, but it is 250V 4.8 - 6.4Amps. I would like more control and I am unsure if this could control a circuit at 120V. Any good suggestions about how to control the electrical without spending more than ~20$?

[Edited on 20-1-2012 by Endo]

Magpie - 20-1-2012 at 14:50

Quote: Originally posted by Endo

... I was able to cut some channels to hold a nichrome element salvaged from an element for an electric furnace.

Just out of curiosity what was the voltage specified for the electric furnace you salvaged?

Endo - 20-1-2012 at 15:01

4.8 KW at 240V. It looked a lot like the element used
http://www.sciencemadness.org/talk/viewthread.php?tid=9705&a...
about half way down to load test. I have four of them in new condition.

[Edited on 20-1-2012 by Endo]

Magpie - 20-1-2012 at 15:07

OK, then you are apparently taking elements that were designed for 240V and applying 120V to them. I don't have any problem with that.

Endo - 20-1-2012 at 15:13

I guess I was just going to be lazy and not run the wire and buy the breaker to hook up 240V supply into my basement. Is it harder to buy/build a control for 16 amps at 120 vs 8 amps at 240?

watson.fawkes - 20-1-2012 at 18:13

Quote: Originally posted by Endo

I guess I was just going to be lazy and not run the wire and buy the breaker to hook up 240V supply into my basement. Is it harder to buy/build a control for 16 amps at 120 vs 8 amps at 240?

Easiest control at this level is an appliance controller, as for a stove or oven. Easy to find as a replacement part. If there's an appliance refurbisher in your area, they'll have buckets of them in their shop. You might have something of a hard time finding a current rating of 16A @ 120V, as such appliances are almost always run on 240V.

You should be aware that your elements are going to run at 1/4 power unless you center-tap the elements you have. If you use the elements at half their original length, you could also wire it as a two-zone oven with two controllers.

Endo - 20-1-2012 at 18:44

I was aware I wasn't going to get that high of a wattage from this element. I ended up using about 3/4 of the length of the element when I wired the furance. I wanted around 1500 to 2000 watts so I wouldnt have to worry about a higher amperage breaker, (it is on a 20 amp breaker now)

I do have a controller from an electric range. However it is typical of most electric range controllers and its lowest setting just after off is 4.8 Amps, then it can be adjusted on up to the limit of the controller. Could I use both legs of the electric range controller to increase the handled amperage of the controller, in a sense put the same phase 120 volts to the L1 and L2 terminals, and run both the H1 and H2 to the element, with the neutral going directly to the other end of the element.

Is there any way to use a pair of 1000 watt dimmer switches in parallel? or will the differences in load handling burn one of them out? Could i put something in the circuit with dimmer switches to prevent them from having the problem with having too much load on them and not being able to evenly balance the load between the two of them.

johansen - 20-1-2012 at 19:18

the controllers from electric ranges are typically thermal circuit breakers with adjustable spring tension, the on time is proportional to current squared, off time is relatively constant. You can pull a little more current than they are rated, but its easy to burn them out.

Short answer is no, you can't wire the dimmers in parallel.

As Watson suggested you could center tap the element you have now, each with its own 1000watt dimmer.
As far as I know, the Triac in a 1000 watt dimmer is probably rated for 20 amps, the 1000 watt limit is more of a thermal dissipation problem when confined to a 20 cubic inch electrical box w/o airflow. the 600 watt dimmers typically have a 10 or 15 amp rated triac.

Endo - 20-1-2012 at 20:25

@Johansen

Ok so In a pinch I could buy a 1000 watt dimmer, dissassemble the case and look up the rating on the triac(s) inside, if it can handle the amperage I then add a heat sink. Otherwise the center tap (breaking the element into two elements) and dual heat zones with a pair of dimmers is the way to go.

Thanks for your reply and help... I will admit that I am not very knowledgable with the electronics.

Magpie - 20-1-2012 at 23:34

If I understand you right you have one shortened element which draws 16 amps at 120V. If you cut that in half and make two equal sized shorter elements you will then draw 32 amps through each element. I don't think you want to do that, do you?
This is what I think you are talking about :
center tap.bmp - 703kB

I think watson was just telling how to obtain the full power of the element using 120V instead of the 240V when mentioning the center tap (correct me if I'm wrong here, watson).

What I think we arrived at was a circuit like this pending the successful use of the 1000w dimmer:
furnace circuit.bmp - 703kB

This will allow you to keep the same 16amps which, as you say, will keep within your wiring/circuit breaker limitations but still give you 2000w.

[Edited on 21-1-2012 by Magpie]

johansen - 21-1-2012 at 02:09

Now that i re-read your post, its clear you have a 16 amp at 120vac element in it now.
--no reason to center tap it then.

I'm not sure what a 1000w dimmer costs these days but i would just go ahead and hook it up.
if you blow the triac you can get a 30 amp unit for $3.00 on mouser.com, or anywhere else (S+H will cost more than the part)

watson.fawkes - 21-1-2012 at 07:39

Quote: Originally posted by Endo

4.8 KW at 240V.

This problem is just an application of Ohm's law. V=IR and P=I^R=V^2/R. If you put half the voltage across a resistor, it draws half the current and dissipates one quarter of the power. The controller has to be rated larger than the current draw, but that current is determined by the load, not by the controller. At the original rating, R = 12 Ω and I = 20 A.

To compute the power draw at half the voltage, V_1 / V_0 = 1/2, and three fourths of the resistance R_1 / R_0 = 3/4, we get a factor of (1/2)^2/(3/4)=1/3, so the power draw is 4.8 kW / 3 = 1.6 kW. R= 9 Ω and I = 13.1/3 A. Therefore, in this configuration, a 10A controller is insufficient.

White Yeti - 27-1-2012 at 08:01

These furnaces look amazing! They make me want to make my own. Right now, I use a wood stove for my inorganic syntheses.
In my kitchen there's an old electric stove/oven that might soon fall apart. That's raw material for an electric furnace.

Endo - 3-2-2012 at 16:40

Well, I found 1000Watt Dimmers for $7.50 each online. It showed up monday and I had time to mess with it today. The Triac inside is a BTA25-600B Looking up the spec sheet it is indeed rated up to 25 amps. A quick try with the back cover of the dimmer switch off had the small copper plate heat sink getting too hot to touch in the course of about two minutes.

I dug through some stuff and found an old heat sink from an Athalon Thunderbird processor and drilled it to mount the triac. Screwed it down with a layer of thermal paste. Soldered some wires to get some working room from the dimmer an gave it a shot. I found an old computer PSU fan and wired it up to blow across the heat sink.

The system seems to work very well. I just ran it for 20 min and the heat sink was barely warm to the touch. I can control the heater down to where the element is just warm, and up to bright cherry red. 4.5 amps to 15.5 amps on the ammeter.

I took a few pics of the furnace and heat sink. Also wrapped the furnace in pink insulation then aluminum foil. Then made a box out of left over ductwork scraps (I know it isn't pretty but it works

) I stuffed the back and corners with more insulation and closed it all up.

watson.fawkes - 5-2-2012 at 08:42

Quote: Originally posted by Endo

Well, I found 1000Watt Dimmers for $7.50 each online. [...] I dug through some stuff and found an old heat sink from an Athalon Thunderbird processor and drilled it to mount the triac.

This is really good practical modification. I blurbed it in another thread on controlling heating mantles. I suggested there using this in conjunction with a dedicated ammeter on the panel.

Could you provide us all the exact brand and model number of the dimmer you used? It would be good to know, if ordering one of these by mail, that it has a triac of the right packaging to make this easy. The RD91 package the triac is in makes the mod much easier, particularly over remotely mounting a TO-220 package or worse yet, one of the PCB-only packages. See the spec sheet for the BTA25-600B to see that it's certainly possible to get unlucky on the device packaging.

Endo - 6-2-2012 at 18:29

The model of the dimmer is LUXDM1000-02L The manufacturer name on the packaging is SecurElectric Corporation.

http://www.homelectrical.com/single-pole-1000w-slide-dimmer-...

The back cover is held on with two screws and comes off easily. It comes with four holes at the top and four at the bottom, so I didn't have to drill to get the three new wires out.

To get the Triac out I used a dremel tool with a cutoff wheel to cut the top off of the pop rivet holding the triac to the heatsink (A small copper plate) and through the aluminum facing. A few minutes with the solder iron got it loose, then just added the wires and pulled them through the pre-drilled holes in the case. Two were really easy because they matched up with some solder joins for other large wires. The third connection was a fairly small patch on the board to hook the wire but easily doable.

Oh I wasn't 100% but from the documentation I could find on the Triac I belive the back plate of it must be grounded. I just drilled a second hole and used a screw to hook a ground wire up to the heatsink.

Hope this helps if someone else wants to try it.

watson.fawkes - 6-2-2012 at 18:52

Quote: Originally posted by Endo

The model of the dimmer is LUXDM1000-02L [...]

Oh I wasn't 100% but from the documentation I could find on the Triac I belive the back plate of it must be grounded. I just drilled a second hole and used a screw to hook a ground wire up to the heatsink.

Hope this helps if someone else wants to try it.

Thanks for the model number. The spec sheet for the triac states the following:

Quote:

By using an internal ceramic pad, the BTA series provides voltage insulated tab (rated at 2500V RMS) complying with UL standards

In other words, the metal mounting tabs on the case are electrically isolated from the triac connections or the device itself. Hooking up a bonding wire to the heatsink is still the right thing to do. The heat sink is a conductor whose potential would otherwise float. It's always right bond such a conductor to ground. This bonding wire doesn't ordinary carry current; it's for conducting fault current to ground so that other protection devices operate.

What has helped me most is knowing there's a 25A triac hiding inside that consumer device. Single unit quantities for that model triac is more than $5, so why not buy the whole assembly? I've been thinking I'll make a general-purpose dimmer box with it with its own outlet, since it would be able to handle anything that the 15A circuit on a typical wall outlet would. My guess is that a similar 600W dimmer unit has a 15A triac inside, for that matter.

Pages: 1 2