Wiseman Ceramics

With the new JD18-JH, I was supposed to get 2.2 mm Kanthal APM elements. At the point where the kiln was ready for shipment, there was a 4 week wait on the APM wire of that gauge, so I opted for the 12 gauge (≈ 2.0 mm) Kanthal A-1.
One of my plans was to compare the lifespan of APM to A-1, using digital images, ohm measurements, and kiln interface software (K.I.S.S.) to record what was happening as each set started to go.
An educated guess was that it would take longer for the APM’s to fail, so I thought that it made sense to get the A-1’s first after all.

I’ve fired several kilns in the past using both “standard” (15 gauge) and “heavy duty” (14 gauge) elements. With either, once the coils start leaning in on each other and bunching up in the corners, their performance declined quickly.
Most of those past firings were in the ^9-10 range, occasionally pushing for a “soft ^11″. After about 15-20 firings, the elements would show the leaning and bunching described above. Past that, I’d average 25-35 firings total before the rate of rise to peak was noticeably slowed, and my firings suffered.

Currently, I’ve performed almost forty ^11-12 (true 90° bend on a self-supporting cone) firings* with the JD18-JH.
When graphed, my rate of rise is still tight with the programmed set point line, and ohm readings taken after each firing remain the same… but hey, pictures are worth a thousand words:

Unfired 12 gauge Kanthal A-1 Elements:

…the same elements, after the 37th ^11-12 firing:

The elements settled nicely into the element holders after the first firing… aside from that, and the oxidation coating on the metal, the elements appear to be in perfect condition.

My Current View on Kanthal A-1 vs. APM Wire:

With A-1’s being less than half (almost 1/3) the price of APM’s, the latter would need to possess at least 3-4 times the life span of the former to qualify the difference in price. One could argue that the downtime associated with replacing the elements factors into the equation; however, if I can get 100+ firings in between element changes, I’ll really have to consider whether I want to replace them with APM’s. My reasoning here, is that accidents can happen with either… For instance, if a bit of glaze, ceramic, or even so much as a few grains of sand fall onto your APM element, I don’t imagine that it would have a better survival rate over any other wire. At that point, replacement costs can really hit home.

*Note: For the record, there are also nine ^04 bisque (I usually use my other kiln for this), and sixteen low temp luster/ glass enamel firings on this element set.

In the spring of 2006, I had an electrical meter connected by a certified technician from the local power company. This meter was set inline between my kiln and the breaker box and was separate from the main meter, so it recorded only the power used by the kiln. During that time I performed firings in the high, middle, and low temperature range to calculate what different firings were costing me.

These firings were performed in a 7 cu.ft. L&L JD230HD with:

• Single phase/240v
• Computer-controller w/ “type S” thermocouples,
• 3″ K23 softbrick (walls, floor, and lid).
• Standard (small) gauge elements… which were not new, but not worn out either. They had eleven firings (four ^04 bisques, and seven peaking at 2345-2355F) total on them when the meter was hooked up.

The meter was zero’d out prior to installation, and I kept track of the energy consumption by photographing the dials as reference.

Firing 1: Peak at 2345 F Hold for 2 minutes + 5 hours of holds b/t 2000 and 1900 F.
Reading: 87 Kw.

The national average for a kilowatt hour was 9 cents,
87kw/h x $0.09 =
Cost: $7.83

_________________________________

Firing 2: “Fast Glaze” program to ^018 (1384F).

Reading: 104 kw/h - 87 = 17 kw/h used.
Cost: $1.53
__________________________________

Firing 3: 270F/hr to 2000F.

Reading: 139 kw/h - 104 = 35 kw/h used.
Cost: $3.15
__________________________________

There were seven firings total, all yielding the same costs (give or take a few cents) per firing.

It would have been nice to see what the difference was in terms of new elements vs. the ones at this stage –and even vs. aged elements (although I rarely let mine get that far gone). But since this meter was on temporary loan from the power company, I was still interested in seeing these readings, and I think they may represent the longest stage in terms of the elements lifespan.

I currently have a meter hooked up to my JD18-JH …you can see those readings as they progress HERE.

Main Kiln Energy Consumption Page.

Disclaimer/Warning:
Automated kilns weren’t meant to be left completely unattended. I fire to a witness cone viewed through a spy hole, so I am there every time it hits peak. Others rely completely on the controller. If this is the case, you can try to Write your own Cone Fire program in order to smooth out some of the variables.

_________________________________

Many are finding inconsistencies in their glaze results from one firing to the next. The variables are vast, but I’m going to focus on three reasons attributing to inconsistencies. The first is that the kiln elements decline with use. The second is that kiln loads vary in weight and how that weight is distributed from one firing to the next. The third is due to heavily insulated kilns that hold too much heatwork following the top end of the firing.

The first thing to understand, is that a cone measures heatwork, not just a given temperature. Heatwork is a measure of heat + time. Time refers to how long it takes to get to peak temperature, as well as any hold maintained once that peak is achieved.
If the kiln is in good working order, it should have no problem achieving the desired rate of rise and peak temperature. But, if your elements aren’t new, or if the kiln is packed heavy, it could take longer than you programmed to get to the intended temperature. If your kiln achieves the programmed temperature over a longer period of time, over-firing will most likely result.

Example: ^10 (true 90 degree bend) equals 2342°F, only if the kiln can maintain a 108/hr rise during the last 150-200° prior to peak. If it takes longer, then the temperature you need to reach will be less. Again, both heat and time must be factored in.

To continue, I’ll need to use my L&L Dynatrol (Bartlett V6-CF) controller as reference.
If you wish to write your own cone fire, program a Vary Fire schedule the same way as usual, except that when it asks for the peak temperature, push the Other button. Keep pushing Other until you see Cone displayed. Type in the cone number you want to achieve (e.g., “10″). The controller will calculate the temperature to hit based upon your programmed rate of rise. Then, if your kiln lags during the firing, the controller will recalculate the temperature based upon the actual rate of rise.

For heavily insulated kilns yielding an over-fired cone, I’ve found that inputing a cone offset helps. Doing this can aid in compensating for the residual heatwork responsible for bending the cone beyond the point you want. Before you do this, you’ll need to monitor a firing to see how much to offset. Also, keep in mind that if your kiln is heavily packed, the ware will hold heat, accounting for inconsistencies as well.

Inputing a Cone Offset: Before you begin inputing a program, hit the Other key until you see CnoS displayed, then hit Enter. Typing “90″ before the desired offset temperature will yield a lower actual temperature, decreasing the amount of heat work achieved. Again, the idea here is that the residual heatwork will bend the cone further (hopefully closer to perfection).

You can only offset the cone measurement by 50°F, but if you need to do more than this, contact the manufacturer, as it eludes to a larger problem. Offsetting in this way will only affect the cone you select. It will not affect your thermocouple readings during other Vary Fire or Cone Fire programs. Speaking of thermocouples, none of what I’ve said here will help, unless your thermocouples are reading accurately. See the post on Studio Kiln Thermocouple Calibrations for more on this.
After reading this, you may begin to understand why I always fire by a visual cone in achieving the correct level of final heatwork. Computers haven’t completely replaced humans yet for a reason! I also prefer a kiln that allows for a fast descent, largely eliminating the effects of residual heatwork.

In February 2008, I was able to get the technician from the energy company to install another inline meter, this time on the JD18-JH electric kiln (for referencing the last meter, see the post: Electric Kiln Actual Energy Consumption).
The meter was set up to record ONLY the electrical consumption of the kiln.

After my 30th firing to ^11-12 and many lower temperature firings, the elements on the kiln look great, and ohm readings from the elements are pretty much the same as when the kiln arrived.

Meter Start Point:

When the meter was installed, it read: 400Kw/h. This will be our “zero point”.

_____________________________

1st Firing: This image was taken during a crystalline firing, immediately after seeing ^11 achieve a 45° bend.

Reading 1a: 434 - 400 = 34 kw/h used.
Cost (at 9¢ a kw/h): $3.06

… following the ^10.5 peak, there was the crystal soak phase of the firing. This involved ramping up/down and holding between 1900-2000°F (total hold times ≈ 4 hrs).

Reading 1b : 456 - 434 = 22kw/h used for heat-soak.
Cost: $1.98

Reading 1c: 456 - 400 = 56kw/h used for the entire firing.
Cost: $5.04

Summary: The total cost of a ^10.5 crystalline firing, including the 4hr. heat-soak costs about $5.00, and the $2.00 it cost to grow the crystals pretty much does away with the notion that crystalline firings are much more expensive than corresponding firings to the same peak.

This firing was done in the JD18-JH. One of it’s design characteristics is to cool out of the top heatwork zone before the perfect cone bend (and the glaze!) can be altered by residual heatwork. One of the arguments against this design was that it would either have trouble hitting the temp’s I wanted, or it would require so much power that firing would be expensive… obviously, neither is the case.

Following the crystal growing holds (heat-soak), the drop through quartz inversion and below was slow and smooth. When I unloaded the kiln, my witness cones looked nearly identical to what they did when I stopped the kiln’s peak temperature hold and began the descent.

Of course, if I’m going to fire the kiln, I may as well put something in it, yeah? …

__________________________________________

Second Firing: achieved ^020 @90° bend:

Reading 2: 463 - 456 = 7 kw/h used.
Cost: 63¢

__________________________________________

3rd Firing: Achieved ^05 @90° bend:

Reading 3: 482 - 463 = 19kw/h used.
Cost: $1.71

_________________________________________

4th Firing: Achieved (another) ^020 @90° bend:

Reading 4: 489 - 482 = 7 kw/h used (Compare to 2nd firing: Man, I like seeing repeated accuracy).
Cost: 63¢
__________________________________________

Related Links:

Electrical Consumption of a JD230HD (@7cu.ft.)

Main Kiln Energy Consumption Page

There are many good kilns available, but choosing one to suit your needs can be difficult. Take note that simply because a kiln boasts a “Rated to Cone 10″, “Energy Efficient” or “Heavy Insulation” label, it does not necessarily mean it is the best choice.  As a matter of fact, that last one can even cause you a good deal of grief.

I believe a good analogy to use toward an electric kiln, is that of a boat.  This may seem like an odd comparison at first; however, consider that both require power to operate against resistance, and since neither has brakes, it is up to the user to compensate for the dissipating momentum. This “momentum”, in the case of a kiln, translates to residual heatwork.Firing a heavily insulated kiln then, is comparable to piloting a heavier and less agile boat –for even after a pyrometric cone achieves the desired bend, it can easily drop well past that perfect arch if your kiln releases heat too slowly.It is also important to realize that automated controllers don’t make this any easier to deal with. A controller can stop supplying power to the elements, but it cannot make a kiln cool any faster than it’s ability to release heat. The user still has to figure out how to compensate, and then program the controller accordingly.I believe the best way to get the kiln that is perfect for your firings is to either build it from scratch, or start with a commercially manufactured shell of the size needed. Whichever choice you go with, the options to consider adding should focus on the need for accuracy and account for the components that will wear out the fastest.The variables below are those that led to designing the kiln I fire.

Variable 1 - The ability to heat:The kiln should at least have the ability to climb all the way to it’s top-rated temperature at 270°F/hr –without lagging. To accomplish this, the kiln needs enough power. I would suggest erring on the side of an “overpowered” design, for if your kiln lacks the ability to ramp fast when new, it will surely be difficult to reproduce results after the elements and other components age even slightly.The correct match between that power and the heating elements must also be calculated. Whether you go with standard or heavier gauge wire, it will have to be factored into a good watt density. As each kiln is going to be different, this subject gets much more involved than this post allows for, so I suggest speaking to a qualified individual (contacts I’ve had good experience with include: L&L, Geil, Skutt, Euclid, Kanthal).APM’s are often said to be the best choice for high temperature work. The long holds (”heat soaks”) used to grow the crystals in a glaze cause “grain growth” in standard element wire, thus weakening the metal. The thought is that since APM’s possess a different structure developed through a sintering process, they aren’t as susceptible to the damage produced by long holds at high temps.Having said this, my kiln currently runs with longer lengths of 12 gauge Kanthal A1 elements. This wire has proven more resistant than thinner gauges to the effects of higher temperatures, as well as the chemical attack occurring from clay and glaze during the firing. These elements currently have over 60 ^11-12 crystalline firings on them, show no visible signs of degradation, and perform with the same efficiency as they did when they were new.-Also, A-1 wire is about 1/3 the price of APM. Beyond power and a good watt-density, the kiln will need adequate insulating properties to retain the heat. Notice that I chose the word “adequate”…

Variable 2 - The ability to cool:Heavily insulated kilns heat well, because they hold heat well. This fact is often used to qualify them as “efficient”. But such a kiln’s Achilles’ heel resides in the inability to release heat at higher temperatures. I do understand that this goes against the current industrial tendency to add more and more insulation, but I hold that the value in a kiln that allows for a rapid diminishing of heatwork after achieving the desired cone value is extremely underrated.In 2006, Diane Creber passed on to me a wonderful collection of literature on crystalline glazes, including some of David Snair’s notes & letters. This information, along with my own correspondence with Derek Clarkson and others, showed a common preference in using small, high-powered kilns capable of completing the first and second phase of the firing (the climb to peak & descent to the first hold) in a short time frame. This, in my mind is a truer example of “efficiency”. I have fired several types of clay and glaze in kilns of various sizes, and there’s nothing like relying on a kiln capable of completing the entire firing program as it is actually set by the user.Many of the more spectacular glazes I’ve worked with require a very accurate firing. With crystallines in particular, my goal is to heat the glaze well past its melting point, possibly maintain a hold to dissolve silica, zinc, and excessive nuclei, then quickly halt run-off by descending into the next ramp. As all of this must occur within a very small window of time, excessive residual heatwork is the last thing you want.Variable 3 - Structural Integrity at High Temperatures:It seems obvious to choose kiln parts that are designed for the temperature and type of firings intended. But this is something that is overlooked often enough.At it’s core, a kiln is simply a strategic stack of insulating firebrick. These bricks, cut to fit and usually wrapped in a metal skin, make up “the kiln shell”.K23 brick is the current standard for most hobby/studio kilns. If you are only going to ^6-^8, then it’s a suitable choice; however, both ^9 & ^10 (as high as 2381°F, depending on your rate of rise), occur at or beyond the recommended range of k23. The hotface, or the side of the brick exposed to the interior of the kiln, can shrink and crack if exposed to temperatures beyond the rating (e.g., k23≈2300°F). Keep in mind that this will occur even if you are only firing that hot occasionally, or if you are firing to a slightly lower temperature and holding there to achieve the equivalent level of heatwork.In designing my own kiln, I selected k25 brick, because I like the option of firing hot… often bending ^11-12.My firings show four important benefits with regard to K25:

• The k25 brick is less fragile than k23.

• The k25 brick performs extremely well from 2000°F to peak without deforming…

• …it then allows for fast cooling from the top temperature down to 2250°F.

• It still, however, releases heat slower through the middle to lower phases (1500-500°F), when dunting and crazing typically occur.

Variable 4 - Accuracy:Along the same line of using brick rated to your needs, consider the ill-advised use of Type-K thermocouples (rated only to ≈2200°F), which still come as standard equipment in many high-fire kiln models. The option of using protective sheaths on Type-K’s can create a lag in temperature readings. My suggestion is to always go with a thinly sheathed Type-S thermocouple for any firings above^6.Refer to the post on Studio Kiln Thermoucouple Calibrations.If you’re replacing a kiln of roughly the same size, go with your notes on how that kiln fired –both up and down. Did one section typically lag behind during the ascent and/or descent? Now would be the time to add insulation to that particular section, or to the floor or lid. Remember to start with the least amount you need however, and add more only where it’s necessary… otherwise, you’ll be right back in the same (big) boat, so to speak.Variable 5-Efficiency & Energy Consumption:An argument against building a kiln as described here, is that the result seems under-insulated, and to require more power.  It may therefore appearing inefficient by modern standards.What I am focusing on here, is a small to medium sized kiln meant for studio use, when reproducing accurate levels of heatwork is important. If your goal is to run several large kilns on a daily or weekly basis, and you are firing glazes with a wider heatwork tolerance, then the few dollars you save per firing is certainly a significant consideration.On that note, and to qualify my view toward energy efficiency, I refer you to the relevant post on the actual power consumption of an electric kiln.Related Links: New L&L JD18-JH

Thermocouple Calibrations

I’ve currently performed seventeen ^11-12 firings with my new kiln, and the elements look and read great, showing no signs of getting tired.
After my 15th firing, I did a test run by programming a 350deg/hr. rise to Cone 9. The kiln managed the programmed ramp to the calculated temperature –no problem.

*As a reference, with my older kiln using the standard run of holders, I was averaging 25-30 high-temperature firings, and noticed signs of element fatigue by my 15th firing. Keep in mind, that I’m doing some pretty harsh firings, so others may get better life.
But an extreme test is often the best test.

Back to L&L JD18-JH Electric Kiln

Ever wonder how much it really costs to fire your electric kiln?

Follow these links:

L&L JD230HD Kiln (@7cu.ft.)

L&L JD18-JH Kiln(@3cu.ft.)

It’s so cold, that I thought I got the coolest crystalline glaze ever… until it thawed.
It’s so cold that i don’t want to joke about how cold it is. Seriously, it’s not funny - it’s really too damn cold.
-How cold?:
I have to warm up my thermocouples with a lighter and set the first ramp to hold at 75°F for 5 hours (instead of setting a delay). Otherwise I get an “Error Code-6“, because the controller can’t read something SO COLD!!!

I’ve tried hanging a “warming bulb” inside the kiln, and also used a small space heater when I was in more of a hurry. But you can’t program a delayed early morning start (when it’s going to be the coldest yet) with any one of those in there –it might smell bad. So I heat the TC’s with a lighter and then program the first ramp with a hold at 60-90°F for “X” number of hours.
The nice thing is that whether I do a delay or a preheat ramp/hold, the new kiln doesn’t lag like other kilns I’ve fired. So my rate of rise is always within a few degrees/hour of what I set (even the -20F night we had here last week didn’t slow it down), and I can pretty much calculate when it will hit peak … I’ve never been off by more than 5 minutes.

This year I’m going to build an actual kiln room (rather than just a walled off area) with a low ceiling and door. I’ve been waiting on the Geil to arrive before I do this…
Once a room is up, I’ll be able to use the space heater the way it’s supposed to be used… I’d only need to keep the room above zero, and I could even warm up the room more prior to when I would be in there loading. Oh, oh… the JOY!!!
Did I mention Wisco gets a little chilly?

Whether you’re trying to achieve an accurate firing via a peak temperature hold or a calculated ramp, it’s a good idea to calibrate your thermocouples (TC’s) before relying on the factory setting in your controller.

Hopefully you have Type S TC’s … if so, you’re not going to need to calibrate them again for some time. It’s pretty ridiculous that Type K’s are used past 2100°F these days. If you have them and plan to go beyond that temperature, you’ll need to check and adjust for “drift” about every 5-10 firings.
No matter which TC’s are being used however, I always suggest using witness cones to judge accuracy on each firing.

___________________________________

There are two ways to calibrate your thermocouples. One way is to fire your kiln, inspect the cones afterwards, and then make a guess as to how to offset your TC’s. This can take anywhere from a few to several firings.   I hear of many who do it this way, and honestly- I’ve never understood why.

A more effective approach is to record what your kiln controller reads during a firing, when you see the cone achieving that perfect bend, then establish the offset from there.

Here’s how:

1- Check with your manufacturer to see what the factory TC offset for your kiln/controller should be set at, then verify that each TC is offset to the number they give you.

Example: L&L Kiln’s factory offset is usually 18 (not zero),
when using the Bartlett V6-CF control board & Type S TC’s.

2- Stack your kiln with stilts and shelves so that an Orton self-supporting witness cone can be viewed through each spy, and place another in the center of each shelf. I also put in glaze tests in addition to cones to get as much information as possible. I prefer not to fire a “light” kiln, as the loss in thermal mass could prevent an accurate result, so I use softbrick and/or extra stilts to take up the empty space.

3- Program the controller for a rise of 108°F/hr during the last 150F prior to peak.

Example: ^10 = 2345°F at 108°F/hr.
So program a rise of 108°F/hr from 2145°F to 2345°F.

4- Set a delay if needed so that you can be there around the time the kiln reaches peak. …Obviously, you’ll want to err on being early.

5- Record the temperature of each TC at the time the corresponding cone bends perfectly. Understand that the readings may be different for each zone/TC, and that they can be offset independently.
The perfect arch (90° bend) for a self-supporting Orton cone is when the tip is level with the top of the triangular base.

6- From here you can figure out if your kiln is under or over firing, and by how much.

7- Your controller can be programmed to offset the TC readings by as much as 50° in either direction. To adjust the TC offset on a Bartlett V6-CF, you will need to calculate the number to offset (also refer to Bartlett’s online manual):

To lower the temperature/heat work in the kiln, start with the factory set point and add the number of degrees to compensate for each thermocouple. This will raise the temperature displayed.

To raise the temperature/heat work in the kiln, start with the factory set point and subtract the number of degrees to compensate for each thermocouple. This will lower the temperature displayed.
To program a number below zero: type in “90″, immediately followed by the number of degrees to compensate… think of the 90 as a (-).
Consider that since 18 is the factory set point for an L&L with Type S TC’s, you should not need to go below zero. If you find that you do, I would suggest contacting the manufacturer of your kiln or controller.

___________________________________________________

After going through this process with my new kiln, I found that my top TC needed an offset of 19 and my bottom at 25. Leaving it at the factory setting of 18 would have resulted not only in over firing, but also in uneven heatwork between the zones (sections) of the kiln. Go here for more on this.

Before reading this, you may find it helpful to review the post on Studio Kiln Thermocouple Calibrations.

_____________________________

While waiting for the modified Bartlett V6-CF Controller, I fired the JD18-JH using the motherboard from my JD230/3 –and it performed great. But now that the new controller is here, I’ve been having some issues getting my kiln to fire to a perfect cone.
Hey- it happens! The thing is, most people fire their kilns without ever taking into account that it may be uneven. For many, it doesn’t matter too much –but when you’re still cooking a glaze that’s already been melted for several hundred degrees, you need all the accuracy you can get.

Once Dave Myers (Bartlett Instruments) informed me that “18″ was the TC offset baseline for an L&L DynaTrol, things became a whole lot clearer.
TCoS:18 still resulted in over-fired cones however, and the bottom zone achieved slightly more heatwork than the top… but at least I have a place to start.

The following information shows how I’m attempting to calibrate the TC readings, using Bartlett’s TCoS function…

Firing 12.16.07
TC offset was set to TC1: 16 , TC2: 16.
The rate of rise during the last 150°F was keyed in: 108°/hr, within a “write your own cone fire program“, with ^10 set as the peak.
Cone 10 was reached in the bottom zone at TC reading: 2335°F.
Cone 10 was reached in the top zone at TC reading: 2342°F, about 4-5 minutes later.
This was not during a hold. It never reached this temperature because the controller’s calculation of ^10 was scheduled at 2351°F, and I skipped step out of that ramp to prevent over firing.

***I was going for a “^10 and a half “, by soaking at 2350°F until the desired bend. So the cones in the following image don’t show a true ^10, but rather what the cones looked like when I skipped stepped out of the last ramp (once my desired level of heatwork was achieved). But one thing they do relate very well, is the slight difference in heat work.****

According to the controller, the rate of rise was maintained at around 108°F/hr.
So if one compares the temperature when cone 10 was at a 90° bend, then zone 1 was too hot by (3°F + offset:16 ) 19°F.
Zone 2 was too hot by (10°F + offset:16 ) 26°F.
And there you have it — too hot and a difference of 7°F existed between the TC1 & 2 readings to boot, easily accounting for the cone measurements.

An educated guess at this point is that an offset of TC1: 19 and TC2: 26 might bring things true and even. I’ll know for sure tomorrow…

I have to add that I’m looking forward to getting the 4-1/2″ ring powered and into the game. The idea for that ring was originally to give me a place to cut a spy hole that you could actually view cones through. This would really help right now, as Bartlett’s limit for a Cone fire program is ^10. Because of this, I’m having to calibrate the TC’s by comparing the controller’s ^10 calculation to a self-supporting ^10, and then continue to hold (soak) to push ^11&12. That’s a problem, because it’s extremely difficult to see more than one cone in that itty-bitty little 1″ spy that electric kilns are typically made with.