Ben can correct me on this, but I seem to remember that he had problems with Nichrome elements burning out quickly on a P1 when connecting them to a PID. It was reason enough that I was happy to find the thicker, electric oven type heaters in the Turbo Oven I use with my Pretty Good Roaster (PGR). That said, I am now wondering how responsive these heating elements really need to be. Dan says he does well with Nichrome wire, and it is very responsive, but I figure that since the idea is to ramp to a certain point, hold it for a while, and then ramp some more, possible in several stages, that it doesn't really matter how responsive the element is to cooling. Heating response time will get over run by the mass of the beans anyway, it seems. I know that if I try to roast a Kg of beans it takes quite a bit longer than if I just want to do enough for an espresso shot. I did a lot of roasts with hot air poppers, and my main concern with them was some chaff getting to the element and causing a hot spot, shortening the life. That never happened to me, and I just retired my poppers. They did not die (other than I had a hard time toward the end avoiding breaking glass chimneys. They either fell over or I dropped something on them.) Anyway, the PGR sees a lot of chaff in its heating element, because it all recirculates around and around. A lot of it burns against the heater, and it continues to work fine. I did bring a spare bread machine and a spare Galloping Gourmet, camping. You just never know. I bet I have roasted a hundred pounds with it (just a guess, could be more) and it still does fine, but I DO love spares! Spare bread machines cost over 30 bucks here in MA. Only five in LHC. Stainless Steel mixing bowls just last forever. PeterZ Catching up on email, camping, here in MA.
<Snip> Concerning longevity. Nichrome oxides. When it does, it loses some metal. As it loses metal the diameter of the wire reduces. As it reduces resistance goes up until such time that it fails. The higher the temperature it is operating at the faster the oxidation rate. And, the more times the wire cycles between hot to cool (delta T) the sooner it fails. Controlling the heater with a on-off percentage cycle timer cycling the heater for a portion of a 10 second cycle the heater will not last very long, a month or two. (The same is true for a PID set to cycle 5 seconds or greater) Controlling the heater with a PID set at 1 second helps a great deal since the heater is cycling so fast that it doesn't have time to heat up very much or cool down very much. The wire will last about 6-8 months. Controlling the heater with a SCR/TRIAC, or other continuously variable device like a VARIAC, will give you the greatest life from a bare nichrome heater. About a year or so. Jeffrey and Doug have done some cupping that says the latter also produces better tasting coffee. I have no reason to disagree with them and will eventually run my heater with a TRIAC. <Snip> Yes, I'm convinced, after a year of use, that bare, coiled nichrome is the way to go if you want to profile roast in a small roaster. Your assessment is incorrect. First, I find no advantage to programming a soak period into a PIDs ramp profile. I don't use that feature. But that's a moot point. Even during a soak period, where you are maintaining a constant temperature in the roaster, the heating element is still cycling on and off for a portion of a second or more, depending on how you are controlling it. This Delta T leads to nichrome failure. However, like I said before, using a continuously variable controller is the way to go and all but eliminates the cost of replacing heaters. <Snip> Did you know? These are nichrome heaters, too. They are called tubular heaters. What they do is encase the nichrome in a mineral filled steel tube. This keeps oxygen away from the heater and lifespan goes up dramatically. The downside is a horrible heater response time. The heater lag is so great that a controller has great difficulty following a ramp. However, it is possible to use them to maintain a steady state temperature in an environment with few external factors effecting the oven. Is use fuzzy logic PIDs to control a bank of 15KW ovens at work using tubular heaters. We have no problems bringing the ovens up to temp, without overshoot, and maintaining plus or minus one degree. When I first made my electric sample roaster I used a tubular heater. It cost $100. I did not like its response time, even when operated by the percentage timer in a manual mode. Sure, you can jury rig an element made for another appliance, but still, they aren't cheap. After it burned out, I replaced it with a $1.35 piece of coiled nichrome. A person can do the math in their head to see that nichrome is the best value. When you add the responsivity factor bare, coiled nichrome is the clear leader. Dan
Hi Dan, Thank you for your response, and your input on NiChrome wire. You have enlightened me on the subject. I did not know that that I was using is tubular nichrome. So far I have not seen that this slower response is significant in the PGR. When I adapt it and make it a CCR the quicker response may be more important. Right now I am happy that the tubular version seems safe in the chaff recirculation environment it exists in. The PGR has no controls other than the thermostat switch which I have found to be useless to determine the actual temperature. It is just easier to use it, but I monitor the TC and vary the ramp time by adjusting said thermostat switch. When I want the element off, I rotate the know until the indicator light goes out, rotating the other way makes it go on, and the heat goes on. When I get to 320 F, I turn the heater off and the roast 'coasts' up another seven or ten degrees. This is the slow response of the element I guess. Not so between first and second crack. If I cut the heater power towards the end of first, it will coast in several minutes, all the way to second. Heater response lag, and exothermic reaction I suppose. The better the roaster is insulated, the more you can see this. Right now, I have used the PGR enough that it is very predictable, and I can roast coffee to suit me and C (she). It would be nice to have *better* control. I would love to super insulate, add more heater, use a larger mixing bowl, and experimentally determine how much coffee can be roasted with a 20 amp circuit, 110 V. Not with my current PGR though. It serves me well. Since I have all the parts, I will need to make a Super PGR, computer controlled, and maybe coiled NiChrome wire. I agree that cost is the issue. The Galloping Gourmet comes with a tubular heater built in. Other convection ovens have the coil wound heater. I have the GG, and it is the most powerful. So I chose it to use. I have lots of nichrome wire, recycled from other heaters, and that may be a next step, but so far I am too lazy to modify what is working well, for me. From what you said if I were going to buy this stuff new, the wire would be the way to go. The way to determine how fast a heater response time is needed would be to somehow evaluate how quickly the beans absorb the heat that is applied to them. With the air popper I found that I could generate 1000 degree F temperature, resulting in scorched beans and lots of divots, but the beans would not be cooked all the way though in some cases. The need to have time in temperature, but how fast do you need to accelerate this temperature increase? I do not know this, but I do know that I can and have applied too much temperature with an air popper. With the PGR also methinks. Dan Bollinger wrote: <Snip> Some folks have been trying to let the roast set at certain temps for a bit. I have been delaying the temp rise by cutting power. This seems to make the coffee taste smoother. <Snip> My cycle time is very slow. Almost a minute off. I turn it off a few degrees before I want the temp to stop rising, and do not turn it on again until I see it level off and hold for a few seconds. <Snip> The slow response helps maintain temp also, while the power is off. <Snip> When doing small batches, less than 70 gr or so, it is easy to roast too quickly with my set up, it *is *more difficult to get a 13 minute roast. That is why the computer would help I am sure, but you may be correct and I may have to switch to wire even with that. Small batches are doable, just need to be cautious. PeterZ Just did 1.4 lb of Harrar with no problem, wonderful coffee, here in Gloucester, or anywhere. <Snip>
Peter, You can see the response time in this test I did. The electric heaters are self-explanatory. You can see why I went with nichrome. I added the gas data over the weekend. While gas responds very quickly, it can never go to zero since you must keep the burner lit by turning it to LO. Danhttp://www.claycritters.com/coffee/heater_response_time.htm<Snip> since <Snip> much <Snip> nichrome <Snip> produces <Snip> the <Snip> assessment <Snip> into <Snip> temperature <Snip> This <Snip> the <Snip> tube. <Snip> great <Snip> We <Snip> made <Snip> I <Snip> math <Snip> unsvbscribes) go tohttp://sweetmarias.com/maillistinfo.html#personalsettings<Snip> unsvbscribes) go tohttp://sweetmarias.com/maillistinfo.html#personalsettings<Snip>
Ahem- The humble toaster uses bare nichrome wire as the heating element, as does the hair dryer. Also true of electric clothes dryers. In electric stoves, the actual heating element is nichrome wire, encased in compacted alumina, in an Inconel or similar protective tube. Industrial U-shaped elements with fins are made for heating air flow in the curing ovens for screen printing. Coaxially in the center is the nichrome wire. Curing temperatures for the common screen printing inks is 325 degrees F. You wouldn't want to get your hair or your clothes that hot. The point is- if you can see the heater glowing, it's up around 1,000 degreesF. You can multiply the life of the nichrome heater element if you can run it, say, 100 degrees cooler. Suppose you wanted an absolute maximum air temperature of 450 degrees F for a fluid bed roaster. You have that now, but your heater burns out annually. That's not bad, but you could extend the Mean Time Between Failures (MTBF) to at least 10 years if you made the heater physically larger, and ran it at 750 degrees instead of 1000. How so, you ask? You van get nichrome heater coils at appliance repair shops, and they probably get them from Johnstone Supply. W. W. Graingers probably has them too. Any shop that handles temperature instrumentation (Omega Engineering, etc) either has this stuff or can tell you where to call. [ The downside is a horrible heater response time. The heater lag is so gre= at that a controller has great difficulty following a ramp.] If you're using a PID controller, it compensates for the hysteresis lags. Cheers -RayO, aka Opa! <Snip> ribes) go tohttp://sweetmarias.com/maillistinfo.html#personalsettings<Snip> -- "When the theme hits the bass, I dance the Jig!" - -Virgil Fox at the Wichita WurliTzer
$100.00 for a tubular heater? They saw you coming! On 7/5/05, raymanowen wrote: <Snip> reat <Snip> scribes) go tohttp://sweetmarias.com/maillistinfo.html#personalsettings<Snip> -- "When the theme hits the bass, I dance the Jig!" - -Virgil Fox at the Wichita WurliTzer
<Snip> You haven't priced industrial heaters in awhile! The heater was about $75, I seem to recall, but the company has a $100 minimum order. What really pissed me off was when it burned out after about 30 roasts. I expected it to have the life of a heater element in my oven. Dan
I have a number of tubular heaters that were shipped incorrectly to me at some point. If someone wants some to play with, I will let a few go for cost (under $10 apiece) plus shipping. I will look up wattage if there is an interest. At 07:24 7/6/2005 -0500, you wrote: <Snip> John Nanci AlChemist at large Zen Roasting , Blending & Espresso pulling by Gestalthttp://www.chocolatealchemy.com/