Help with HM and Keg? Dale/Andy


 

Mike Wallace

TVWBB Member
Guys I'm having trouble with using the HM to maintain 225 on my Keg. I have Ralph's rotodamper. I tape all around the bottom vent with aluminum foil ductwork tape. Top vent at 1.5. I use a 15" round 3/4" thick pizza stone lowered into the firebowl for a diffuser. I fill the firebowl pretty full, above the holes, and light with looflighter in the middle top of the pile. I've tried Bryan's egg settings. I've tried Dale's keg settings. Have the fan set to max 50%. On all the time (not at max only).

1. For startup I light the fire, put the diffuser in right away, and put the cast iron grate on the upper setting. I set the pit temp to 150. It fires right past the 150, and keeps going to over 200 and starts leveling off around 210. When it levels off, I bump the pit temp up in 5 deg increments until I hit 225.

2. Once I am around 225, the first time it drops below 225, the HM stokes the fire, it overshoots to 228-230. At this point it has to shut down to 0% until the temp drops. This takes 15-20 mins. At this point the fire goes out and when the damper finally opens and fan kicks in when dropping past 225, it's too late. I have to take everything out and relight.

How are you guys lighting your fire, and what are you using for a diffuser? I'm wondering if the thick pizza stone is too much.
 
The keg is about the most insulated kamado around, it will take a LONG time to drop temp no matter what you do. At start up, the HM doesn't know that - it needs time & data to 'learn' how to control the pit. PID settings wont matter in your case (yet, you gotta be 'in control' before PID does anything). You need to make ALL the air flow through the servo, but not let the servo go full closed. 5%, 10%, whatever it takes to keep the fire from choking. If you close it 100% off, the fire is supposed to go out...If you're only overshooting to 230-240 on the 1st time up to temp & it takes 1/2-1hr to drop back down, no big deal - that big hunk o meat will cool it off quicker.... PID doesn't stop oscillations, it makes them smaller than they would be naturally....

I only have a fan, but I can't run <250* on the keg w/HM, the natural draft through the fan is too much. At 250-275, it runs rock steady with or without heatermeter (HM does better long term as the fire conditions change). I use a lot of foil tape & I've added permetex around the lower damper seal. I've pounded the slider in so that it fits tighter, etc. Rather then wire my HM4.0 for a servo damper (I have servos & a 3d printer), I just run at 250-275. Meat gets done faster & tastes the same. I don't use sugar in rubs & don't use much sauce, so there's no burning & I can't really taste the difference 25* makes.
 
Also, maybe leave the CI grate out until it over shoots the 1st. QUICKLY open & add it. That will suck up some of the 'excess' heat & speed up the reaction. HM will eventually see that it is easy to overshoot & slow to come down, then start to control it better. It just takes time to learn that.

Also, hit the button to activate lid detect BEFORE you open the pit if the lid is going to be open for a while. Open pit + fan rush = HUGE overshoot. Only thing you can do after that is wait. Ask me how I know!!
 
Yeah, I've pounded the intake tracks closer, sealed both the lower and top vents with permatex red. My gasket is a little hard in spots, and leaks smoke where it sits on the rivets. I've ordered a Rutland flat rope gasket to replace it like they do on BGEs, permatex it to the lower body over the rivets instead of the lid.

Maybe I should try adjusting the lower limit for the servo, so when the output goes to 0, the damper is actually still open a very small amount (size of a pea) so a little air still gets through the fan/damper and doesn't let it completely snuff out.

Could also be the lump I'm using, Basques Sugar Maple from Costco here in Canada. It's not as hard as other lump, so it burns hot which makes it harder to control low temp.

My next experiment is to use a large 3lb 6" coffee can. Guys on the keg forum have done that, they added a pipe to the bottom side of the can, and the pipe goes over to the ATC. Three bolts slightly above the pipe hold a grate in the can above the pipe. Then you fill the can with lump and wood chunks. The can keeps the fire contained and smaller.
 
I wouldn't re-calibrate your RD, trying to fool the HM is a fools errand.
For low and slow around 225F you need a very small fire to hit your target temp. This means you should be careful how many coals you light at once, becuase a large fire will overshoot 225F easily. I also suggest that at 225F with a very small fire you need very low air flow. If you leave the blower active when it kicks on and starts to blow it will actively stoke the fire up to be larger than what you need. I suggest you change your fan settings over to ON AT MAX ONLY and allow the convection air to control your pit rather than forced air. If you for some reason insist on using the blower during the cook at such low temperatures then you will need to set the MAX speed for the blower very low.
 
I wouldn't re-calibrate your RD, trying to fool the HM is a fools errand.
For low and slow around 225F you need a very small fire to hit your target temp. This means you should be careful how many coals you light at once, becuase a large fire will overshoot 225F easily. I also suggest that at 225F with a very small fire you need very low air flow. If you leave the blower active when it kicks on and starts to blow it will actively stoke the fire up to be larger than what you need. I suggest you change your fan settings over to ON AT MAX ONLY and allow the convection air to control your pit rather than forced air. If you for some reason insist on using the blower during the cook at such low temperatures then you will need to set the MAX speed for the blower very low.

Ralph, I think what he means is that once you overshoot on the keg, if the RD is at 0 opening, by the time the temp drops enough to cause the servo to re-open, the fire is out. It needs a bit of air, which will make the temp drop slower, but will keep the fire going. He needs to 'fool' the HM a bit on startup until the temp settles in a bit, otherwise the HM gets ahead of the fire & causes overheating issues. A keg is a different beast vs ceramic, it's extremely well insulated. At 700* dome temp, I can bear hug the lid, except 2-3" from the lid opening & 1-2+ from the top wheel. Try that on a BGE... I've killed it at 800* from a burn off at ~11PM and it was still warm to the touch at 7AM the next day in my 50* garage.
 
I agree with Ralph, start with a very small fire. I run into the same thing on my uninsulated UDS, it takes forever to bring the temperature down at start up if i pour to many lit coal into the fire basket.

The heatermeter cant be fooled, there is nothing in the heatermeter that can be fool. It does not learn anything it only reqcts to low and high temperatures. The only real adjustments you have with the heatermeter will only adjust when the heatermeter will react to the high or low temperature and how much it will react to the differences, using the PID adjuments.
 
Something you may want to try for #1 in the original post is to set the "Max startup fan speed" to something lower than the normal run speed. The original desigh there was to have a HIGHER max fan speed during startup so it got going faster. My dad was having trouble with overshoot on startup and was doing the same thing you are, but eventually changed his max startup speed to something like 30% and it come up to temperature and just stops with only a small overshoot.
 
I think he was setting the pit at 150 at first to try and avoid the building of the fire too big, which I think could be cured easier by reducing the startup max on the blower like Bryan suggested (and then you can just set the pit at your actual target temp when you fire it up).

I still disagree with calibrating the RD so it is open a bit at 0%, if you set the fan on at max only you will not get the rapid overshoot that requires the damper to clamp down long enough to put out the fire. If you let the damper run the pit with natural convection (blower not running at all after the target temp has been reached, on at max only) you overshoot period should be much less, and with a damper on the input you can run the top vent more open and the pit can breath a bit from that. The HM and servo damper do a good job of building up or scaling down a fire to be the right size for the target temp, but it is best to get a fire of the right size going in the first place and avoid long periods of stoking or choking.

So give the "on at max only" mode a try... I used to set the temp lower when stoking the pit to game the overshoot but since I got the roto damper and on at max only working together I have no need for that any more. I just set the target temp, the blower will stoke the pit until the temp is nearing the target temp and then it turns off (PID settings will control when the HM starts to back down). When the rush of air from the blower stops the temp tends to shoot up a bit, sometimes overshoots other times not. (This is one of the problems caused by forced air control, it stokes the pit at the same time it forces large amounts of cool air into the pit that make it seem artificially cool, then when the air flow stops you get the overshoot.) With on at max only mode the damper will generally clamp down on the air flow for a few minutes at most and then opens up a trickle allowing the pit to settle right on the target temp nicely using the natural convection forces of the fire.

Here is a graph of a test I did recently with my prototype RD3 using the on at max only setting:
RD3Graph.jpg


On this cook I simply lit 1/2 a weber starting cube under a pile of coals, when the starter cube was burned off I closed the lid (don't like to trap those fumes in my grill). The fan was blowing until the first dip, in this case the pit temp leveled off short of the target, the servo damper opened wider and wider to try to hit the target, eventually it hit 100% and one more little burst of fan blowing set me right on the target. This was the first test burn with the RD3 using stock PID settings, after a few cooks I tailored the RD3 PID settings a bit and she lands pretty much right on target with only a short window of overshoot of a couple degrees if that....
 
Last edited:
The keg can be a real challenge as it holds heat so well it will not drop the temp quickly and my take hours to drop back to 225 after a 15-50 degree overshoot. It is insulated so well, 3 pounds or so load of lump has lasted me over 26 hours at 225 without adding lump in fact it still had enough unburned lump I could have gone at easy 5 to 6 hours more.

I have my servo set fully closed at 0 and full open at 100%. I have fan set to 30 max only on at startup or max not sure which don't have it on-line right now to check.
I use a small piece of a sawdust wax starter cube under one piece of lump on the top center of my large pile of lump. I usually set the top vent at 1.5 to 2 depending on the wind and desired temp. For 275 - 300 it need to be open 2.5 to 3 depending again on the wind.

The bottom vent is sealed with foil tape all the way around. I did not permatex my keg at all. I never adjust my PID I only adjust the top vent to help control temp based on the wind.

Different brands of lump have so far not seemed to make much difference in holding the temp study. I light the starter then close up the keg with the vent full open and, with the temp set at 200. I wait at least 10 to 15 minutes then set lid open alarm to close the servo and add the meat. Close the lid then set the vent to 1.5 -2.

I let the lid timer, time out, I have the lid timer set to 10 minutes. I watch the temp awhile and wait for the temp to start leveling out close to 200 then change to 225.

I then set pitdroid app alarms for high and low pit temp and got to bed, Most of times there are no alarms, at all. During the day I watch it on my phone until it is done and time to check it 12-18 hours latter, never opening the keg, if I open it to foil I start by setting open lid.

I never use water and use a foiled brinkmann foil lined water pan as a defuser.

I have only had a fire go out once and that was from using way too many small pieces of lump and it got too blocked with ash, this was before I used the heatermeter. Make sure you start the base of your lump pile with larger pieces of lump the try hand loading the lump so it does not compact from being dumped in, use small pieces of lump on the top as they burn the ash can fall through the spaces between the large lump on the bottom falling through the grate. Too many small pieces can be a real problem causing the fire to go out. To many wood chucks can cause to much flare up and uncontrolled burn.
 
Last edited:
Bryan, thanks I'll try the new startup max, with a low setting, I haven't upgraded the software yet.

The fire I have been starting is small. Looflighter on one piece of lump at the top of the pile. However I've been having the fan on 50% or more which probably stoked it into a large fire on startup.

I also haven't been sorting the lump, just dumping it in.

Dale, what are your latest pid settings?
 
B =O P =14 I = 0.004 D =5
FAN SET VOLTAGE MAX 22%, START UP 22%, ON AT MAX ONLY.

If you put the start-up fan speed higher every time you adjust the temp it will run at that higher speed until it reaches it for the first time this will also cause it to overshoot.

Part in sorting is with the large pieces on the bottom you not blocking the holes in the grate. In the BSK forum I had heard about sorting it making a big difference.

Are you using wood chunks or chips?
I prefer the chunks and I never soak them. have read it just creates a lot of steam, the combustion air is what controls the burn rate and smoke.
 
Last edited:
Chunks, no soak.

What fan do you have? I'm using the standard recommended one from the HM hardware listing last year.

How are you connected to the keg intake? I have Ralph's rotodamper, connected to an intake plate with electrical conduit which has about a 1" ID.
 
Same here Ralph's Rotodamper with the stock fan same connection.
I used 1"conduit connector mounted to a small piece of steel plate inserted into the bottom vent track, then sealed with tape using hard drive magnets on vent plates to help seal them to the keg.
 
Last edited:
Dale, what are you putting the Brinkmann pan on? I'll see if I can try your settings and sorted lump tonight, but my thick pizza stone diffuser may be slowing my response even more. If my next trials don't work, I may need to try a thinner steel pan diffuser instead.
 
Still doesn't work. Fire is out. Set startup to 10%. Set temp to 175, once it overshot 175 set to 200. Overshot 200 by 28deg. Once it overshot 225, set the temp to 225. The 20 minutes it took to come back down to 225, the fire had been put out.

The keg is too insulated/efficient. It requires such a small fire to maintain 225. I need to set the damper to not fully close. Doing that IS NOT fooling the HM. Running the HM with no damper, just a fan, is similar, at 0% the fan is still letting draft air in. Is that fooling the HM? No. If I set 0% to have the damper open just a little, say 1/4" round ball, maybe that is enough to clamp down on the fire, but not put it out.

Or, I need to stop using aluminum foil tape on the intake adapter plate. That will let some idle air in, instead of having the intake sealed off 100%.
 
Last edited:
got threaded hooks at home depot drilled 4 hole in lip of water pan and and hunk from grate. Are you sure fire is out, mine is air tight and takes hours for the fire to go completely out. Where do you have your temp probe positioned mine is hanging on the back of lid thermometer, I found little difference between that and the grate temp. Are you sure your lump is good and not wet or some thing once it is lit it should be very hard to put out.
what model keg do you have? mine is the 2000 with upgraded 4000 cast grate and upper wire rack.

with the damper not fully closed you may get runaway temp. the heater meter with just the blower I had to keep the top vent almost shut about a toothpick diameter open to keep form over drafting and over shooting the temp
 
Last edited:
try a different lump I think your lump is screwed up. I would love a twenty minute to drop temp mine usually takes an hour or more to drop 25 degrees, but never had it cause the fire to go out.
 
You need to tune PID settings....and shouldn't have to go through the whole set at lower routine then raise to target. If you get your PID tuning down the HM will start to back off from 100% (and fan will turn off) as it approaches the target temp, then only slightly overshoot for a bit and settle at which point the damper opens up to let air flow.
And if your grill is that efficient put only a small amount of coals in there. My "fauxmado" leaks like crazy and a small load of coals can do a couple cooks, so no need to load it up and light a bunch of coals because a small fire and small bunch of coals will keep a low and slow cook going a long time.
 
Fire was out. Dropped to 155. Might be the lump, I keep it outside, but 100% protected in a steel cabinet under a granite countertop, under a covered porch. But maybe the humidity through the paper bag is enough to dampen the lump.

I re-lit, and removed all the tape. That didn't work, hit 245 and still climbing. However I re-lit by putting a weber cube in the bottom under the fire grate. May have lit too much of a fire.

The pizza stone diffuser is my next culprit. I pulled it out, and it was radiating a ridiculous amount of heat. I really think it is slowing down the reaction even more/too much. I cooled it off under running water and put it back in. Now it seems to be reacting like it should. Maybe with the high heat latency pizza stone, my setup likes to run with a fire from the bottom instead of the top?
 

 

Back
Top