It looks a lot like node-red for linux based SBCs like raspberry pi, and I think they do android now as well.
Hey Matt, do you think instead of the servos youâve linked before that your mixer will work straight up with an electronic throttle body? Almost all the ones Iâve fiddled with use 5vDC.
Im not sure how you would control it with code. I dont know how they work.
A servo controlled with Arduino has code pre made so that you can just instruct the movements with angle positions. I can tell it move one angle postion at a time for instance.
Servo.write ( Last Position + 1); or (Last Position - 2): Or I can make go directly say a 45* angle possition.
The Throttle bodies I think have stepper motors and I have no idea how they zero the position or how to code them.
That makes sense, Iâm assuming if I wanted to pair a mixer valve with a servo then Iâd just print or make an arm to push on the mixer valve with a rod?
At least then in an âall else failsâ event I could hook up a choke cable to it.
No think man you have a 3d printer you can make anything!! No you integrate the servo into the valve and couple it directly to the angle plate.
What size valve you need? I have the 3/4 NPT version for 400 to 500 cc engines and now the 1 inch NPT version for the V Twins. You will have to forward an email address and I can then send you the STL files.
Hi Cody, many older throttle bodies works exactly like a servo, iâve tested with a cheap servo driver/tester, and it worked fine.
I belive this is often true with the type with a âslenderâ electric motor, and spring return.
I would check if i can find the one i used, to see which model it was.
Yeah if they postion using the same PMW single they should work without modification to code.
I guess the easiest way is to build the controller and plug it in. Its either going to work or it wont. If it works and we need to flip the direction let me know and Ill make that happen.
Otherwise any servo will work that is based on a 0 to 180* range. Even though in code Im thinking in angles, the code converts that to a PWM signal for us.
I think it depends on the throttle body, apparently some use servoâs and some others may use steppers. stepper just move in steps that are like 1.8 degrees apart or some multiple of that.
How difficult do you think it would be to make a MAP sensor based mixer? One motor of course, but somehow using two identical MAP sensors for the readout.
I guess youâd make the servo somehow balance the air flap until the MAP sensors are reading the same? Using vacuum readings youâd likely have a faster startup on woodgas since you wouldnât need to wait for an O2 sensor to warm up.
Also if anyone has some good tutorial links for programming please share.
The AEM ( Bosch) sensors have zero wait they output instantly they have built in heaters.
Yeah dont re invent the O2 Sensor is the best way to do this it works and no need to re invent. You are looking at months to a year for that type of code dev if it will even work at all. Vacuum driven systems are not accurate eniough correct AFR is correct AFR and the only way to determin that from a sensor is with an O2 Sensor.
If you were going to tap it into the exhaust would you put it next to the original one right before the cat converter? It looks like temu has the "oemâ bosch sensor kits. The sensor is 23 bucks, but they have another kit with the fitting you weld on and the gasket ring things. I didnât check aliexpress. But it avoids messing with the existing emissions system.
You need the AEM kit or the off brand version, They come with the sensor in the kit. The controller needs the Lamda signal out for its input so it knows what to do.
Apparently someone has at least some code for it along with a schematic. Which it is telling me I posted this in '15 lol
Yeah Im good I just got a massive migrain reading this post and thinking how I would have to spend years re-inventing this code to make it work properly. Im good bro.
Off to take some pills god my head hurts! lol
Besides that you need the guage so you can see the DRO and know what you are doing and to monitor the controller when its in automix mode. Im sure you can code that but then you are buying a readout and then mounting it, wiring it etc. Its not worth your time when you can buy the whole kit for less than $100 bucks and it will work seamlessly.
Matt youâve written before that the servos use 180 degrees of motion, wouldnât that make a valve rotate that much as well? Or are the degrees factoring in an internal gear reduction from inside the motor?
No because if you remember I limmited the code to a 90* window. (45* - 135*) Code can only tell the servo to move within that window and can not go outside those parameters.
So when you install the servo to the valve. Make sure to power it up with the controls and using the manual control knob turn it all the way counter clockwise. The servo is then preset to a closed state. Likewise install the valve with the angle plate closed. When you fully turn the knob fully open the full 5 volt signal is then mapped from 45* to 135*.
The AEM lamba 5 volts signal is mapped to a delay variable. Actually two sets of delay parameters.
The leaner the higher the voltage signal; the richer the lower the voltage signal. The lamba analog 0-5 volt range is mapped to 1035 bytes. The window for stoich mixture is somewhere between 600 and 700 bytes. So I map 700 - 1035 bytes to a millisecond delay function and likewise 0 - 600 bytes for the rich condition.
The closer the range gets to Stoich window the higher the delay is set in each senario.
So if we detect a lean condition for instance.
If ( O2 Senser Reading is = > than 700 Bytes)
Mixer Servo = (Last postion -1); // tells the servo to move 1 increment in negitive direction.
Variable Delay; // Tells the controls to wait for delay parameter time to pass before exucutating
code again.
So that last line " Variable Delay" has to pass before it will execute this section of the code again. When its way out of parrameters we want it to respond as fast as possible, however we need to slow it down as it gets closer to the safe window otherwise it will overshout and just hunt going back and fourth. So this is why it has this variable delay. As it the servo moves closer to the safe window the delay between single possition servo movements increases. This also helps with sudden temporary spikes.