Blue Sky Stepper Damper thread - Breakout from RaspberryPi + LinkMeter blue sky


 
Regarding moving to a larger arduino platform - I saw that you did a great job with defining the pin numbers in the header files... I thought maybe it was as simple as redefining the header file pin definitions to new ardunio platform?
 
I just spoke with a buddy of mine who does 3D industrial design professionally. He has agreed to design the butterfly valve that we want in exchange for a pork butt and brisket :wsm:

He asked me to figure out what I want and he'll whip it out for us, give me the 3D models for it, and then we can go to Ponoko or somewhere to get one made.

So now we just need to decide what we want.

For my smoker I use a 1-1/2" PVC pipe to connect the fan to the smoker. I can easily use a valve such that it will just fit inline with PVC fittings. The valve would be a butterfly valve controlled by a servo, or could have lever put on it and controlled by a solenoid. I'd have a return spring put on the valve so that it is closed when everything is off.

If 1-1/2" is too big for everyone else let's make it smaller, I'd prefer having it made so that others can use it. Let me know what you think.

dave
 
OK here is another thought . . . bear with me its an odd one.

For my smoker I have to gate off the intake significantly to ensure the temperature reached by the natural convection of the smoker is ~25-50F below my smoking temperature. Then the HM uses the fan to keep it at the proper temperature. So my first thought is to make a damper to do that gating automatically, but . . . .

What if we just dumped the fan all together and instead used the damper to control the natural flow of air to maintain the proper temperature? We could re-purpose the fan output to control the servo instead.

Bryan, how difficult would that be in the software? I assume that the blower output is a PWM output. We could change the software (configurable of course) to control the servo instead of the blower. Would need to validate that the IRL510 response time is fast enough to control the servo, and some other things. So what do you think? Crazy or reasonable?

dave
 
OK some more out loud thinking. . .

We could use the blower pin to pull a 10k resistor low to make the control signal. We could pull 5V from the button header, and a ground from the power jack, or the serial port header.

Looking at the code it looks like we could add the capability to do the blower, or convert the analog write of the fan speed to the position we want the servo to be in and write that value. Refresh is handled automatically in the servo library.

The servo library adds 850 bytes + 6*(8*#servos) of memory to the build. I think we have that much memory left last time I flashed.
the library can be found here: http://www.arduino.cc/playground/ComponentLib/servo

EDIT: Ok after reading the new library any pin can be used to drive the servo. So in theory we could use an unused pin like pin 24, as thus leaving the fan circuitry unmolested.
 
Last edited:
Yeah that's kinda what I was talking about on the previous page. I think ideally you'd just adjust the damper-- just like a person would do if they were watching it. I don't think the servo library is needed at all, controlling a servo is not much more code than PWMing the fan.

The only reason I didn't go with that originally is the hardware aspect of it. Building physical things isn't my forte, so designing and building a servo controlled damper control that works and isn't a complete piece of garbage seemed a little out of reach. If we can come up with something, I'm game for making the software to do it.

The 1 1/2 inch pipe sounds about right though, for the fan control at least. The recommended blower would fit right into that. For the "pure convection" option, I think something larger would be in order. I could be wrong though, it would require experimentation.

I would think the easiest thing to build (convection) would be a box or tube with the top open and a hinged lid on it. Then the servo would open and close the lid. The lid could have a gasket or something so when it is pulled closed it blocks the air completely.
 
Last edited:
I've got business next week out of the country so I'll be out of touch mostly. When I get back if we can figure out what we want built, my buddy can design it quite easily in solidworks, give us the 3D model and we can get one made. Maybe we can get Dave to get his buddy to print us one :)

So be thinking of what you would like it to look like.

dave
 
D Peart said:
Well imagine that! Some one has already done this:

http://www.geekwithfire.com/Hardware/TheDamperBlowerunit/tabid/64/Default.aspx
Click to expand...


Looks nice, and easy to build. I imagine the cost difference of the servo would be absorbed by the fact that you can use non-blower style fans easily. My fan was like $18, but I had a whole box of 12v 'computer' style fans.

I'd build one of these, but I think my smoker is way too 'leaky' to really gain much. I'll just have to be sure not to over fuel it.

I've finally confirmed my first rPi based HM smoke will be next saturday. I'm excited! Doing pork ribs.

Keep in mind these servos can pull 1W easily just to hold torque.
 
Yeah wow look at that! Wow that even pre-dates my starting this project.

You can use regular computer fans with the existing HeaterMeter system if you want to save money. The issue is just the amount of static pressure flat fans can work against, which is an order of magnitude lower than a blower motor. A 30CFM computer fan might only blow 1CFM or 25CFM through a grill depending on the amount of obstruction which also varies over the duration of the cook.
 
Bryan Mayland said:
The issue is just the amount of static pressure flat fans can work against, which is an order of magnitude lower than a blower motor. A 30CFM computer fan might only blow 1CFM or 25CFM through a grill depending on the amount of obstruction which also varies over the duration of the cook.
Click to expand...
I believe that, in this case, the fan is used to effectively close-couple the damper valve/gate to the system, not drive the flow so much. I imagine a properly designed manifold would make the fan unnecessary, but might be difficult to implement and, at least, cumbersome.

I've contacted the Geek with Fire to see how his concept is working.
 
I priced what it would cost to 3D print a control valve at 1-1/2" and it is way too much money. I then got to thinking that we could use a 1-1/2" piece of PVC pipe, drill a hole in the center of it (may be difficult to do exactly), and then just have the butterfly piece made. It would be a disc shaped piece with a hole through the center to align with the holes in the PVC pipe. You would use a piece of small rod as the axle. Then all we would need to be manufactured is the valve itself, which is quite small.

For this piece we could put a groove around the outside edge to hold an o-ring to give a perfect seal, or we could just size the valve to fit snugly in the pipe. Not sure which method I like, but I'm leaning towards the groove for the o-ring.

What you ya'll think?
 
Here's a question... Did we decide how to drive this? Are we driving this by external control circuit coupled with blower output, or are we looking to use an output pin? We seem to be running out of output pins...
 
My first attempt will use a servo to control the valve with no fan. On my smoker I can easily hit the temps needed for smoking without a fan. I'll power the servo with an external power supply or maybe a battery, I'd like to see how long a couple AA could drive a servo with essentially no load. Then use the blower output as the servo control. I'll have to check the voltage levels for the servo to make sure nothing gets fried, but I think that would be the most straightforward way to do it.

If the servo can't handle the 12V, then I'll have to modify the board. I think the easiest mod would be to blue wire the one unused output and use that as the control and just run a wire to the servo, still using the external supply to power the servo. Other options would be to make a simple control board to convert the blower 12V output to whatever the servo wants. Lots of options there.

dave

Dave Casazza said:
Here's a question... Did we decide how to drive this? Are we driving this by external control circuit coupled with blower output, or are we looking to use an output pin? We seem to be running out of output pins...
Click to expand...
 
Ok, so I'm not an expert at building a control circuit, but I can code - I see a couple of options here, after you evaluate the first option of building a control circuit that doesn't require an extra pin.

I know Bryan isn't a fan of the mega (I'm sorry, Bryan), but I'm in the process of proving/prototyping the pinouts are 100% compatible between the uno and mega. I'm working on an option with that to use the next PWM Digital Out pin (it starts at Digital Logical Pin #14 and goes to #22) on the mega to drive a servo, and keeping the existing control circuitry in place. PIN_SERIALRX (PD0), TX(1), BLOWER(3), LCD_CLK(4), LCD_BACKLGHT(5), ALARM(6), SPI_MOSI (11) and SPI_SCK (13) are PWM compatible across the uno and mega, so this should work.

If the mega doesn't work with the existing codebase, I know we have built-in circuitry to do an alarm, and we have significant investment in keeping PD6 as an alarm. However, how many of us are within several feet to hear an alarm? The whole point of this was not to be tethered to the smoker... which implies the alarm system might be better implemented via software/text message/etc, and not hardware, freeing up a pin for a damper solution.

Anyhow, just my crazy thoughts on this, and my apologies for suggesting something that might be unpopular.

Dave
 
Finally got back to this project. Had a trip to Taiwan, a 1 week backpacking trip in NC on the AT, and work got in the way :)

I just sent my initial sketch to the designer to get it drawn up. My plan as of now is to just build the butterfly valve. It will be sized to fit inside a piece of 1.5" sch40 PVC pipe. This will allow the industrious to build their own setup with pvc pipe instead of paying Ponoko to print up a piece for them. If you need to learn how to drill a hole centered on a pipe google it (v-block and a drill press).

The valve will support a 1/8" axle. There will be a place in the center of the valve to attach a bolt going all the way through the valve, perpendicular and offset from the axle. Two small washers on the bolt will then clamp the axle to the valve. I imagine the assembly to be insert the valve into the pipe, slip in the axle, center valve and tighten bolt.

The axle will be connected to the servo via a "horn". The horn will be constructed using a small piece of 1/8" inner diameter brass tubing (found at your favorite hobby store), bent on a 90 degree angle and epoxied over the end of the axle. This horn will then be used to connect to a matching horn on your favorite hobby servo. The servo can then be glued using silicon glue (holds well and is removable) to the pvc pipe after careful adjustment.

When I get the 3d files back I'll post them and you can see what I'm talking about :) Hopefully this approach will make it very easy for people to make their own. I plan to have the stl files so that you could just print the pipe with hole and servo mounting area as well, but the cost could be pretty high.

To use the valve you will need to place two 1.5" pvc couplers, one on each end, then put it onto your air intake somehow.

The software will need to support a "trim" function so that you can tell it where "closed" is, and enter servo info so that it know how far to open the valve.

that's it for now,
dave
 
Bryan,
Would it be possible to use the wireless connection for the HM to wirelessly control the servo? I'd put a batter on the servo, a small PCB and AVR to drive servo. that way I could use the same HM in servo mode for fan mode with really no hardware modifications.

And then we add the pit and temp probes to the wireless and now my HM doesn't even have to be outside with the smoker!

dave
 
Yeah certainly. My only concern would be the update frequency. The output is adjusted every second and if I start transmitting every second, then I'll probably start colliding with other 915MHz transmissions. Not that it would mess us up, but more likely that infrequent other transmitters would start getting running into problems. Would you still get as good results if the servo were adjusted only every 10 seconds or so?
 
Bryan Mayland said:
Yeah certainly. My only concern would be the update frequency. The output is adjusted every second and if I start transmitting every second, then I'll probably start colliding with other 915MHz transmissions. Not that it would mess us up, but more likely that infrequent other transmitters would start getting running into problems. Would you still get as good results if the servo were adjusted only every 10 seconds or so?
Click to expand...

I think a 10 second update would be fine. When I make changes it typically takes minutes to see the change anyway.

dave
 
D Peart said:
Finally got back to this project. Had a trip to Taiwan, a 1 week backpacking trip in NC on the AT, and work got in the way :)

I just sent my initial sketch to the designer to get it drawn up. My plan as of now is to just build the butterfly valve. It will be sized to fit inside a piece of 1.5" sch40 PVC pipe. This will allow the industrious to build their own setup with pvc pipe instead of paying Ponoko to print up a piece for them. If you need to learn how to drill a hole centered on a pipe google it (v-block and a drill press).

The valve will support a 1/8" axle. There will be a place in the center of the valve to attach a bolt going all the way through the valve, perpendicular and offset from the axle. Two small washers on the bolt will then clamp the axle to the valve. I imagine the assembly to be insert the valve into the pipe, slip in the axle, center valve and tighten bolt.

The axle will be connected to the servo via a "horn". The horn will be constructed using a small piece of 1/8" inner diameter brass tubing (found at your favorite hobby store), bent on a 90 degree angle and epoxied over the end of the axle. This horn will then be used to connect to a matching horn on your favorite hobby servo. The servo can then be glued using silicon glue (holds well and is removable) to the pvc pipe after careful adjustment.

When I get the 3d files back I'll post them and you can see what I'm talking about :) Hopefully this approach will make it very easy for people to make their own. I plan to have the stl files so that you could just print the pipe with hole and servo mounting area as well, but the cost could be pretty high.

To use the valve you will need to place two 1.5" pvc couplers, one on each end, then put it onto your air intake somehow.

The software will need to support a "trim" function so that you can tell it where "closed" is, and enter servo info so that it know how far to open the valve.

that's it for now,
dave
Click to expand...

I run an Auber temp controller on my UDS barrel smoker (55 gallon) I had an issue with it and returned it for service, while it was gone I ran some test runs with a wide open 1.5" air intake (I have six 1/2" exhausts in the barrel lid) temp was from 250f to 275f...so 1.5" might be a bit limited for a smoker that big, might need to run a two of them :-). But for smaller stuff 1.5 might be great. Current interest seems to be in more and more "hot and fast" where temps run 325f or so instead of the traditional 225f.

Here is how I plumbed a 1.5" "waste arm" into my drum, this chrome plated brass tubing stuff comes 1.25" for lavatory and 1.5" for kitchen stuff...and there are existing off the shelf ways to couple it with PVC. That setup there is exactly as I ran it to achieve 250f to 270f with a full basket of Kingsford Blue briq.

SfVCl.jpg


The waste arm is a fairly nice part for like $8 at Home depot. It was not at all difficult to cut it to the length I wanted with a 4.5" grinder with a 1/16" cutoff wheel.


Bill
 
Last edited:
You must log in or register to reply here.

 

Back
Top