INTRODUCING: the "Roto Damper"


 
Yah, I understand, but I wanted to elaborate the reason why a separate ground was used for the probes, and why it might not be needed on the HMv4.2 (yet to be determined). My current HM is v4.2.4 and I jumpered the Servo/Blower GND pin on the CAT5 jack to a spare pin on the CAT5 jack and use that for the probe ground, I get no noise at all over a 50"+ CAT5 cable thats been out in the weather for more than a year, so I would say this works fine.... and a big "HATS OFF" to you is in order for such huge improvements on the HM in recent months!

EDIT: And BTW, I've been experimenting with the new Startup Blower feature and found it to work outstanding on mid to high temp cooks, but for low and slow on my Fauxmado it still causes overshoot even if I set the low speed max between 10-30%. To hold 225 my Fauxmado requires very low air flow and any amount of fan blowing seems to push it into overshoot.... So I would still kinda like to see the "Fan on above XX%" feature some day. I think I had this I wouldn't need to change my HM configuration when I switch from low and slow to high heat cooks...
 
Last edited:
Bryan Mayland said:
I believe Mike is asking about where on the HeaterMeter board to use for ground. Ground is all over! I'm not sure if you're on 4.1 or 4.2, but 4.1 I'd use the GND connection on the probe header, and 4.2 I'd use the GND on the barrel input jack. Just whichever is closest.
Click to expand...

Thanks, that's all I was looking for. 4.2
 
OK, I've been lurking here for a while. I'm a man of few words, but I need to say how great this damper is that I got from Ralph. I had no idea i could achieve such reliable and smooth temps without constant monitoring, and tweaking of vents. Thanks Ralph!
 
Thanks a bunch for the compliments! The roto damper does work really great, but it would be nothing without the Heater Meter and all the tweaks Bryan was willing to do to the hardware and software to make the damper function as well as it does. Speaking of tweaks, the new smooth servo motion on the latest firmware release is a huge improvement, no more jerky hyperactive servo like it had been initially. Anyone that is using any sort of servo damper should look into installing the software updates if you haven't. The RC filters on the HMv4.2 boards help the servo action a lot as well, because the pit temp reads more steadily and therefore the HM makes less adjustments of the servo...
 
I finally got around to re-working my Roto Damper to integrate the external Thermocouple amp and panel mount TC connector this weekend. I used the TC amp board from the HMv4.1 modified to run the most recent TC amp circuit from the HMv4.2 (which I detailed here) I made a case to hold the TC amp and connector which mates with the wiring cavity on the Roto Damper, here is a pic of the parts before I mated them together:
RotoDamperEXTTC.jpg


Here is a pic of the parts mated together:
RotoDamperTC.jpg


Here is a pic of my new Roto Damper sitting next to my Heater Meter in the sliding back case I designed for the HMv4.2:
RotoDamper+HMCase.jpg
 
Last edited:
While I was putting the arrow and lines on my new Roto Damper I figured I would detail what I have found to be the easiest method to get them attached.
First attach the 3 lines and the arrow to small pieces of tape, shiny side down.
RotoDamperArrows.jpg


Next, apply a small amount of super glue to one line, try not to let the glue spill over onto the tape. Immediately affix this line to the Roto Damper body in the center, try to lay it in the exact right place the first time so you don't smear super glue all over the Roto Damper, which can look ugly. (but can be cleaned up with Acetone, super glue remover or fingernail polish remover)

Now attach the (already calibrated) Roto Damper to your Heater Meter, set the HM to Manual Fan Mode, then set the HM to 50% fan speed. Now apply a small amount of glue to the arrow and affix it to the Roto Damper Barrel so it points to the center line you attached previously.

Next set the HM to 100% fan speed and attach a line where the arrow is pointing, then set the HM to 0% and attach the last line where the arrow is pointing. Let the glue dry for a few and then remove all the pieces of tape.

Here is pic of my Roto Damper with the lines and arrows taped in place:
RotoDamperArrowGlue.jpg


Here is a pic of the completely assembled Roto Damper after the tape has been removed:
RotoDamperOB.jpg
 
Well, I guess I jumped the gun on the shared ground... I fired up the pit last night and when she got going I had all sorts of jumping around going on with the probes as the servo and blower did it's thing. As soon as I unplugged the servo and blower the probes leveled out, so it must be the rushes of current from those devices running down the same ground wire the probes are using that is causing the probes to flake out. I will have to go back to a dedicated probe ground I guess, and will update the above diagram when I get around to it...
Here is a pic of the graph showing the probes freaking out:
SharedGndNoise.jpg


The blank spot in the graph is when the roto damper was disconnected completely (when the thermocouple amp is disconnected from the HM the reading will float upward like you see there, the thermocouple was not connected during that rise). After I disconnected the CAT5 cable I pulled off the wires for the blower and servo and reconnected the CAT5 cable and the probes were stable at that point, so that seems to prove that it is the current from those devices causing the issue.
 
Ralph,

On the panel mount jacks you provided me, that go in the damper body, the solder tabs are inline instead of offset like the through hole mount ones that go on the HM board. I found a schematic online but want to confirm, the tab 1 closest to the probe is the probe wire, 2 (middle) is ground, and 3 (rear) gets no wire.

It's a little different than the through hole mount, where 2 is the rear tab, and 3 is the middle side offset tab.

?
 
Last edited:
The front tab (nearest to where you plug the probe) is ground, the middle tab is where you connect the probe lead. (the back tab is the switched leg) BTW, you can use the back (switched) tab to connect a thermistor to get a more realistic ambient temperature than when you put that thermistor inside the HM case with all that heat in there... If you plug a probe into the jack where you have the thermistor on the switched leg the probe will function instead of the thermistor...
 
I checked with my multimeter, and the middle tab is the ground (common). I measure 0 resistance between tabs 2 (middle) and 3 (back) with no probe inserted. With a probe, I measured the probe resistance through tabs 1 (front) & 2. That means 2 is the ground (common).
 
No, that is not correct for the panel mount jacks. If you check continuity between the threaded panel mount (where you plug the probe) and the pin nearest to it you will see they are always connected. THIS is the ground, otherwise when you screw that jack to a metal chassis you would be shorting it out. The center pin is the TIP, the tip is shorted to the third pin when there is no probe inserted via internal switch, when you plug in the probe that switch opens and the back pin is disconnected. So, the pin nearest the front is gnd, the center pin is the tip, and the third pin is connected to the tip through the Normally Closed internal switch.
 
Recently while adding the thermocouple jack to my roto damper I experimented with using a shared ground for the servo, blower and probes cause I thought perhaps the RC filters on the probe lines would clean up any noise. Unfortunately I turned out to be wrong on that, with the shared ground the probe readings would jump around quite a bit when the servo moved and blower kicked in, must be current rushes going through that ground wire freaking out the ATMega.
So I went back to a dedicated ground for the probes and everything has been running smooth as silk over my 50"+ CAT5 cable again. In order to use the thermocouple over the CAT5 cable the thermocouple needs to be on the roto damper end, if you send the raw thermocouple output down the CAT5 cable there is a large temp offset that varies with the ambient temperature, with TC plugged directly into the amp (on the roto damper end) I don't have this problem.
Here is an updated diagram of suggested wiring for the CAT5 jack for HMv4.0-4.2 with probes and/or thermocouple:
HMv4.2.4_CAT5_Wiring.jpg
 
Question that I'm posting here because Roto-damper's seem to be one of the most effective mechanisms at closing off a smoker used here and I know a lot of you with roto-dampers go "servo-only" for low & slow.

Basically I'm noticing in my servo-only setup that the heater meter tends to spend a lot of time swinging between 0% (or very low) and really cranking things open. Before I had this glorious contraption I was pretty decent at manually managing temp by finding the right "dialed in" vent settings. I'd think that a PID would eventually tend to do the same. It seems it's always working to try to get to 0% and only coming up with it has to.

I'm wondering if there are some PID tuning options that might make it want to dial in a bit more and oscillate a bit less. IE: instead of bouncing between say 0% and 50% I'd think it would get closer to 25% and hover around there (I know this isn't exactly right but you get my meaning). I'd think this would lead to the most stable temps overall but maybe I'm wrong. Admittedly I'm no PID expert, just wondering if this is something you've noticed with roto-damper as well or if you have tunings that result in smaller curves vs. spikes..

Here's an example with the default PID tunings:

jhXHco4.png
 
You definitely need to start diving into PID tuning. Ideally, it should find a sweet spot and hold it there.

Your graph should look more like this:

 
JIves said:
What is yours tuned at? Just curious, also is that fan control or servo only?
Click to expand...

It's in fan + servo, but at 30% open on this damper (the rotary one that Tom designed ), there's not a whole lot of air flow.

B: 0, P: 3, I: 0.008, D: 8
 
You must log in or register to reply here.

 

Back
Top