Electronic Controlled Automatic Choke Control

[NITROPIXIE]

Been playing with a well known DIY ECU and have been thinking about impleting an electronic choke control using an idle stepper motor. I know you only have to pull a little wire but just makes it that little bit more fun. It would use a temperature sensor to assess how much choke is required and you program the processor to however many steps you require the choke to be open.

I have drawn a diagram on how to use the stepper motor. The motor has a plunger which is pushed in or out as the motor turns.

Not sure how to supplement the software as yet but will look into it some more unless anyone has any ideas.

[Funky Diver]

The other option could be Servo's, like the kind found on Model Air planes / cars, cost effective too, although I wouldn't be sure on how well they would implement on a car.

Presumably, and excuse my ignorance, it's Out Loud thinking... there would also be some way to operate an LED when the car is up to temperature for the manual choke to be removed. So car get's to temperature to not need choke, illuminates an LED indicating less than optimal performance out of the range of the tuning map.

Sorry, I feel like I'm throwing spanners in the bucket here

[brentp]

Not at all- your ideas are great! There are idle air control devices that are controlled more like stepper motors. Personally, I like the simplicity of the PWM air valve.

[Dean924s]

Rheostat to a "5th" injector triggered by the grounding of the coil. .. . ok bad idea.. . . My mind works in some really weird ways.

[NITROPIXIE]

Been thinking about this some more. And have come up with 2 solutions which would require possibly 2 different hardwares.

One would be the use of a servo as Funky Diver suggested to operate a choke on a carb.

And the second is the use of a PWM idle control valve for mechanically injected fuel systems, such as MK1 and Mk2 golfs, and some other earlier VAGS. Basically anything which had an earlier auxilary air valve or similar for an idle valve which was controlled mechanically by the temperature of the engine. With a 5th injector output option.

I was thinking (maybe over thinking) they would need a temperature sensor for sure, possibly an ignition cranking input for 5th injector and making it control the idle at a set rpm rather than a set amount of choke/extra air (although this maybe over complicating things).

I suppose this would be a niche corner of the market to produce for, especially the second suggestion and the first idea maybe good for those with bike carbs which have a choke option.

[Funky Diver]

I've also been thinking about it for my bike carb implementation on an 89 8valve VW engine.

I'm still exploring the relationship between engine temperature and emissions (ie Lambda) as this would be the another answer to controlling choke rather than rely on temperature alone. There is of course the time it takes to heat up, which would mean that we're talking a correlation between heat of the engine and O2 emmissions.

[NITROPIXIE]

Ok so i was bored at work and drew (in pen sorry, as i have no design software) an initial circuit for my idea. It has managed to incorporate both a servo controller and a PWM controller by the flick of a switch on the circuit board, so you can have one or the other. It also has 2 digital high current sinking outputs to control a 5th injector and some sort of replacement for a WUR (Warm Up Regulator).

Anyone know of a good valve for isolating high pressure (10bar) fuel using an electronic valve??????? Eureka, how about a fuel solenoid off a nitrous system. Normally closed without 12v, ideal. Don't know why i didn't think of this before its been bugging me for weeks!!!!!!! That's the WUR sorted.

Inputs so far are only 2, one is with a thermistor for engine temperature (CTS), which can be run in parallel (or shared if you like) with a megajolt using the same sensor but not a gauge. The other input is a cranking input, mainly for the 5th injector scenerio but otherwise not required for a carb with a standard choke.

Ok so i have my rough circuit design, now I just need to workout:

which processor chip to use (thinking Atmega8 at the moment due to it being a popular microprocessor, is powerful enough),

improve my voltage supply circuit to one similar to a V4,

provide a highish current 5v regulator to power the processor and servo

And finally try to incorporate all this into a current V4 case to keep it tidy (but may need different ends on the casing)

So then peeps give me your ideas on:-

Shall I incorporate a pot to adjust the temperature at which to turn on/off the idle system for quick easy changes (ie winter/summer)??? Aswell as a serial/usb connector to intially set it up and fine tune.

or shall i give it an rpm input so it can stabilise rpm to a set point so it would idle at a constant 1200rpm when cold and slowly bring down rpm as it warms up, making it smart in away and so does not open a prefixed amount at everytime at a certain temperature. OTT or do it for the challenge????

What to call it..... Idlemate???

Give me your thoughts people please.

Funky diver, I'm not understanding why you would want to hook it up to a lambda sensor??? On ecu's the lambda sensor isn't used for a set amount of period until it and the engine is warmed up. Unless your wanting to control the idle at a set AFR using a wideband?? As far as i am aware the first fraction of the choke being open just lets more air in to increase revs, its not until you pull the choke out further does it start to richen the mixture. Correct me if i am wrong??

Ryan

[Funky Diver]

I should clarify...

Choke on > engine starts getting up to temperature > lambda should start picking up readings by this point (if it's a heated one - bosch state they are ready to start taking readings from 7 seconds, let's assume 30 is a decent point at which to take readings from ) > AFR should be picking up rich and starts pulling choke off (fnaaar) in line with temperature readings

Rather than Choke on > temperature reached > choke off

So the process is gradual based on two readings rather than just one, in effect trying to maintain a "rough" equilibrium between a decent ball park AFR reading and temperature. Make sense?

I'm possibly over complicating things, which has been known to happen, but the way I see it is that it's a bit more holistic of the way the engine is running than relying on temperature alone.

From my exploration so far (I'm using ATMel for the schematics I'm working on, prototyping with Arduino Nano's, but probably shifting to Mega's for the pins outputs) it's not overly difficult to do and would tie in nicely with LCD screen displaying the info.

[brentp]

The reason why I lost interest in the LCDs as I'm viewing Android based device as an awesome display/control alternative, tethered with bluetooth. But that's not to say a dedicated LCD display isn't a cool option.

So I've been developing a test board for the SAM7S processor with a bunch of ins and outs (analog in/out digital in/out PWM outs/ timer/RPM in's ....) & bluetooth enable-able. This might be able to be at the heart of what you're looking for. Easily programmable with C or.... Lua scripting (Yes, I said Lua, same language people use to customize War of Warcraft MMORPG )

It's on the Blogging backburner for now, need to talk about Megajolt V4 Rev D first!

[brentp]

^^^ The board I speak of is originally designed for two purposes- a "sensor interceptor" - designed to be inserted between sensors and ECUs- to trick ECUs into thinking they're talking to different sensors, or modify stock ECU behaviour. The 2nd purpose is to be a general purpose test board for us!

[Yvan]

So then peeps give me your ideas on:-

Shall I incorporate a pot to adjust the temperature at which to turn on/off the idle system for quick easy changes (ie winter/summer)??? Aswell as a serial/usb connector to intially set it up and fine tune.

Yes.

or shall i give it an rpm input so it can stabilise rpm to a set point so it would idle at a constant 1200rpm when cold and slowly bring down rpm as it warms up, making it smart in away and so does not open a prefixed amount at everytime at a certain temperature.

Yes.


This is interesting, hope you finish this, I'll test it for you

[Funky Diver]

Thought you might appreciate this NP

http://www.arduino.cc/cgi-bin/yabb2/YaB ... 1281989497

[NITROPIXIE]

I have been going through this to understand the code alot more, its been a while since i have done anything like this and i felt like i have had to learn it all again. But its slowly sinking in following this http://www.earthshineelectronics.com/fi ... alRev4.pdf Been reading it none stop for the last 2 days.

I'm beginning to think i can do the idle control and display all in one. And if i can pulse width control smiths gauges then it means we can get temperature information from one sensor and spread it to differing display outputs.

Then why not ditch the whole dial unit in a dashboard and have a computerised screen with differing dial patterns, which can emphasis different information, e.g. rpm, speed.

One step at a time though.

[Funky Diver]

Seems to me that we're looking at very similar projects, except I was being creative (ergo lazy) and leaving the choke control to you

If it helps, I have an enquiry with a guy who's making nano-mega's, uses the same programming as the 2009 (deumidueyhueygoofy) but has more inputs and outputs to use.
Personally I'm looking to use Pug 407 guages in my car, which should be fairly doable, unless I come across some nicer ones

[NITROPIXIE]

So i thought i would start off with a block diagram. The intention is to have a basic operation and then a more advanced operation depending on the type of system wanted but all available on the same board with some slight software changes. The inputs could be gained from a megajolt unit as per spookie techs rpm counter or individual inputs in parallel with a megajolt could be used. This system could also be used without a megajolt system with an original distributor setup so remains fairly flexible.

Option 1 Basic carb control.

This would require the servo output and CTS as an input. When the CTS is at a certain value the servo will move in relation to this value. Easy!!

Option 2 Advance carb control.

This will require the inputs of option 1, plus an rpm input and a MAP or TPS input. This time the required rpm is programmed in relation to temperature. E.g. 1200rpm when cold, 1000rpm when warm, 800rpm (cold idle off) when fully warmed up. The TPS or MAP input is used to tell the micrprocessor that the car is being driven and so not too control idle whilst being driven until it goes below the MAP/TPS threshold. A throttle switch will also be able to achieve this.

Option 3 Basic CIS idle control

This system does away with the mechanical idle control system and replaces it with a programmable electronic version. In its basic form it uses the 5th injector, PWM idle valve and a fuel solenoid outputs. Inputs used are the ignition switch and CTS in its basic format.

When starting from cold the 5th injector is operated during cranking and for X amount of seconds after cranking to aid starting. If the engine is warm then the 5th injector will not operate to prevent flooding. Threshold can be set in realtion to the CTS. The 5th injector could be pulsed in order to limit amount of fuel injected depending on temperature.

The PWM idle valve is operated on a similar basis to the servo in option 1 and the fuel solenoid closes after a threshold of CTS.

Option 4 Advanced CIS idle control

This uses the rpm input to keep a constant rpm as per option 2. Some CIS thorttle body's have an idle switch built in so this could be used otherwise a MAP/TPS threshold to tell the microprocessor the engine is no longer at idle.

Further options could be Smiths temperature gauge control using PWM, which i believe could be achieved so you don't lose your engine temperature gauge by using the CTS.