Automatically actuated grill block - Mk3

[Daox]

True, it is not closing all the way. That was my oversight. The blade was printed size on size with the frame / opening. With the snapped connections slightly bulging, it hits before it closes completely. The next version will have some clearance designed in!

I'm not overly concerned about the push rod. First, it is really quite short. As its shown now, the length between control arms is about 1". It will be shorter in the final version. Also, when the grill block is shut, the wind force should equal out between the upper and lower faces since the pivot point is in the center. At this point in time, I have not offset it. Thus it should require very little holding power when fully closed (or fully open). I'm also not sure what I'll use on the final version. I just had a paperclip laying around for testing.






I worked on the grill block design over lunch today and I think I have it pretty much finalized. I was able to move the servo a bit closer to the pivot point. This will leave a lip for mounting around the perimeter with some overlap. Unfortunately, I changed every single part on the assembly in some small or large way, so I have to print them all over again. To print all these pieces takes a good 7 hours or so, and 4 separate runs on the printer.





I'm not exactly sure how I'm mounting it to the car at this point. I suppose that is the next step. It will need to go in the lower grill opening. I'm hoping to create a grill block similar to Darin's Insight grill block. It looked very nice and I should be able to cut a hole in it and screw the grill block right to the foam.





The other solution is good 'ol coroplast like I've done with so many cars. It wouldn't be too hard to bolt the grill block frame to the coroplast. It might be a bit on the flimsy side though.

[gone-ot]

TRY it "first" before you print another!

Tis better to LEARN what the original parts CAN / CAN'T do than to make "educated" guesses about their pro/con performance.

[RoadCyclist]

This may be too late, but have you considered making the flap smaller? Not only would it place less stress on the servo, it would also prevent excessive open-close cycles caused by a large amount of cold air entering the cavity.

[gone-ot]

...or, alternatively, use TWO flaps geared together in opposite directions: as one flap rotates UPWARD to open, the other flap rotates DOWWARD also opening...and vice-versa.

[Daox]

Okay, lets do some more testing!

I don't have a blow dryer. The only thing I have is a box fan. I put the fan on max and put the block in front of it (actually off center of the fan where it throws off the most air). While it was running its 'sweep' program. It actuated just fine. It didn't even appear to be struggling at all. So, I decided to hook up the ammeter to the circuit again. This time instead of reading 60mA most of the time and occasionally 70mA, it jumped between the two a lot more. So, it was definitely drawing a bit more power.

Then, just for kicks, I set the grill block face down on the desk. As it opened up, it lifts the entire assembly and lets it back down. While doing this, the ammeter peaked at 90mA, with the normal draw being lower than that.

Finally, I put a normal pliers on top of the assembly (weight of almost a half pound). While the servo struggled to lift it, it did in fact lift it. While doing so, amperage peaked at 260mA. While these tests are in no way a direct correlation to how it will be used in its actual application, I think it shows that the servo is not as wimpy as we originally thought. So, while I am still guessing, I am pretty confident the servo will work fine on the road. The paperclip push rod also held up admirably.

[LittleBlackDuck]

Originally Posted by Daox

Okay, lets do some more testing!

I don't have a blow dryer. The only thing I have is a box fan. I put the fan on max and put the block in front of it (actually off center of the fan where it throws off the most air). While it was running its 'sweep' program. It actuated just fine. It didn't even appear to be struggling at all. So, I decided to hook up the ammeter to the circuit again. This time instead of reading 60mA most of the time and occasionally 70mA, it jumped between the two a lot more. So, it was definitely drawing a bit more power.

Then, just for kicks, I set the grill block face down on the desk. As it opened up, it lifts the entire assembly and lets it back down. While doing this, the ammeter peaked at 90mA, with the normal draw being lower than that.

Finally, I put a normal pliers on top of the assembly (weight of almost a half pound). While the servo struggled to lift it, it did in fact lift it. While doing so, amperage peaked at 260mA. While these tests are in no way a direct correlation to how it will be used in its actual application, I think it shows that the servo is not as wimpy as we originally thought. So, while I am still guessing, I am pretty confident the servo will work fine on the road. The paperclip push rod also held up admirably.

Don't forget that these servos are designed for model planes that fly quite fast and have large aerodynamic forces on the control surfaces. They still move very quickly in that use case.

I reckon that you should file the outside of the door slightly to get the fit better and let it rip!

Simon

[gone-ot]

Now that you've got "real-world" test data showing you that it all works...try it on the car! Hold off on the "2nd Gen" redesign until you've gathered data about what *really* needs to fixed or improved...then, do it.

[RoadCyclist]

There is a good bit of coding to be done first. I imagine it will be difficult to find and access the temperature sensor.

[gone-ot]

Junk yard...or, suitable automotive "ambient air" sensor!

Unless, he plans on controlling "on" the coolant temperature.

[Daox]

Yes, the coding will also need to be done. I don't think it'll be too complex just to get it working. After its working I can think about improving it. The first step will be open before e-fan turns on, close around the thermostat temp.

I am planning on tapping directly into the coolant sensor signal on the car. It shouldn't be hard to find, just one of the many wires behind the glove box I'd assume. However, I'm not exactly sure how to proceed as far as reading the coolant temp sensor signal with the Arduino. From what I've read its just a thermistor. So, to get a voltage reading from it I believe I'll have to create a voltage divider (unless you guys know of a better way, I am no electronics guru) so I can get a voltage reading from it.

I do have this resistance chart for the temp sensor. It looks like the range we're looking at is between 250 ohms and 300 ohms. That seems pretty tight, but again I'm not electronics wiz.





I'm also contemplating how to control the open and close set points. Part of me thinks a 20 turn pot would work real nice for adjusting things. The other part says throw a laptop in the car and adjust the code.