State of Rex hacking?

[Phate]

Thanks Phate for the assistance, A further question, if you have run the battery right down low.

Re the REX operation,
When down to the lowest level as in 6% SOC indicated, where there is effectively no draw in the battery,
Is the 65-70 mph available, OR is that only when both ghe REX and the battery are contributing to the drivetrain?

Just trying to determine what the REX'S power alone is by reference to it's performance when battery is depleted

At 6% SoC the i3 drives exactly the same way whether the REx is running or not, you don't actually get into power restriction until you run the battery farther down to 2-3%. That 65-70mph speed is basically what it can "keep up with" without the SoC dropping - it's somewhere between 19-21kW continuous output, which is used by both the drivetrain as well as the HVAC. Power restriction limits the car to like 40mph or something like that.

[Woodfie]

Thanks for that knowledge, helps understanding what can expect from the REX.

[Woodfie]

Inlet manifold with specific small volume , to be more LPG friendly, than a large induction chamber of gas with combustible mixture, even if uncompressed

Throttle body has common or garden watering look.



Next to print a magnet timing disc, for hall probe sensors
And learn how to fire the coil overs.

[jrbe]

I assume the cylinders are 180° opposed. You should be able to run them wasted spark and tied to the same trigger if that's the case. That could also mean a very simple single trigger on the trigger wheel to get started. Basically it always fires some amount before top dead center, even on the exhaust stroke.

Most newer coils are a 5v logic trigger. You can test with a stack of aa batteries or a DC power supply and tapping a switch. Likely needs something like a 2k pull up to 5v and the switch is normally closed to -. Should for on a quick press on switch release.
Some coils have a dwell "safety feature" that fires the coil when it sees too much dwell and doesn't care about the end of the fire signal (when it would fire normally.) Can end up preignition, detonation, broken starter (normally,) broken engine. Not sure if those coils are setup that way though. Something to watch for if you're not controlling dwell well. You'll likely see ignition timing jumping around at low rpm / cranking if it's an issue.
Newer coils just won't fire if they over dwell, they disapate the energy without firing. It can be confusing because the dwell is typically only a few ms. It's hard to tap a few ms pulse and it seems like they just don't want to fire. I'd guess they are this style.

Not sure of your long term plans but as far as the intake manifold goes, that's likely tuned to the sweet spot of the highway generation setting. You'll probably lose a bunch of internal combustion engine power without the intake resonance / length.

An optical pickup could work well if you have a sensible way to mount it and the pickup.
Hall sensors can be triggered from any ferrous metal if they have a magnet built into the hall sensor. Industrial sensors usually have this option. I've used littlefuse 55505 or 55075 sensors for things like this.
Could likely even do a reed switch for basic testing.

If the manual throttle and log valves / dwell issues are too much of a struggle to make good progress, pickup an Arduino mega. You can run speeduino on it with some bodge wires / a breadboard.

[Woodfie]

Thanks for the helpful input.
Yes cylinders are 180, so have today printed a disc, after purchasing some magnets.. have some hall sensors that will check if suitable, will put sensors at 180, one up and one down, with the magnets one N facing, one S facing
. The coils are BMW, looked at some of the numbers to see if any spec info, but haven't found anything yet.

[jrbe]

Those ignition coils look too small to me to have built in igniters / drivers. How many terminals do they have?

[crasbe]

Those ignition coils look too small to me to have built in igniters / drivers. How many terminals do they have?

According to the schematics, they have three terminals: +12V, Ground and Signal, which is low-side driven by the REX Digital Motor Electronics.
So you'll probably need a strong N-Channel MOSFET to drive the ignition coils.

[jrbe]

According to the schematics, they have three terminals: +12V, Ground and Signal, which is low-side driven by the REX Digital Motor Electronics.

Do you have a picture of this or a link?

So you'll probably need a strong N-Channel MOSFET to drive the ignition coils.

"Signal" in an ignition coil typically means a logic level signal. It could be a bad translation too.

[jrbe]

A kind person shared the info. This is a cut down version of the coil wiring,

Up top is 12v. Brown is ground, likely to the cylinder head / valve cover. White for cylinder 1, white with green for cylinder 2 are the pulsed - to dwell & fire the coil.

So as crasbe mentioned, these coils will need a coil driver / strong Mosfet for them to fire. No worries about dwell besides cooking them long term.

[Woodfie]

Those ignition coils look too small to me to have built in igniters / drivers. How many terminals do they have?

They are Three terminal..
Brown earth center,
Red /Green to the right are common, so likely 12v.
White/Green on one cylinder, White on the other

[Woodfie]

Thanks for sharing info and schematics, Exactly what I was looking for

[Woodfie]

A kind person shared the info. This is a cut down version of the coil wiring,
Screenshot_20240205_090049_Samsung Internet.jpg
Up top is 12v. Brown is ground, likely to the cylinder head / valve cover. White for cylinder 1, white with green for cylinder 2 are the pulsed - to dwell & fire the coil.

So as crasbe mentioned, these coils will need a coil driver / strong Mosfet for them to fire. No worries about dwell besides cooking them long term.

This is Great, What would be a strong Mosfet 3 A ?

[jrbe]

A typical coil driver is a bip373 but they are getting hard to find.
https://www.diyautotune.com/product/coil-driver-kit/

http://www.megamanual.com/ms2/bip373_datasheet.pdf

It's been a while since I've done these things. I'm not up on what's a commonly stocked coil driver. Most are moving to coil on plug with built in drivers. You could look at using coil on plug coils but a lot of them are dwell time sensitive and it might be tough to find one to fit.

https://www.mouser.com/c/semiconductors ... &instock=y has a few through hole options. I don't think typical mosfets last very long driving ignition coils.

You can hit up a junkyard to get a coil driver too. Bosch 211 coil drivers are 4 channel and common in late 90's to mid 2005? European cars. Get the pigtails as well if you go this route.

[Woodfie]

A typical coil driver is a bip373 but they are getting hard to find.
https://www.diyautotune.com/product/coil-driver-kit/

http://www.megamanual.com/ms2/bip373_datasheet.pdf

It's been a while since I've done these things. I'm not up on what's a commonly stocked coil driver. Most are moving to coil on plug with built in drivers. You could look at using coil on plug coils but a lot of them are dwell time sensitive and it might be tough to find one to fit.

https://www.mouser.com/c/semiconductors ... &instock=y has a few through hole options. I don't think typical mosfets last very long driving ignition coils.

You can hit up a junkyard to get a coil driver too. Bosch 211 coil drivers are 4 channel and common in late 90's to mid 2005? European cars. Get the pigtails as well if you go this route.

Some Rex ignition Progress,
An adapter to pick up the spline of the resolver, and 3D printed magnet disc, loaded with the magnets.
Have the magnets and sensor giving a clean switch on check before mounting.
Will check with scope once rotating, for waveform.
Have Established engine rotation is Anticlkwise from resolver view, I.e. suction on the intake.
Plug removed to set timing position. Figure doesn't matter whether the adjacent cylinder is ahead or behind 180 °, as is wasted spark method .

So onto coil, have 12v on, and the WS 16 connection is at 12v,
So confirming its a two wire system, operation the third wire is coil ground return ,
I.e. a normalish coil.
Mr Kettering used a condenser, not seeing such on this diagram, looks like resistor of some size..
Any thoughts?

[jrbe]

From the diagram, I see:

Position 3, +12v.
Position 2, high voltage (spark) ground.
Position 1, coil - / coil fire trigger.

Digital ignitions don't have condensors that I've ever seen, those are to keep points life up. If you're using an ignition rated driver I don't see why you would need one. If it's not an ignition rated driver a condenser should help it live a bit longer.
It should not need a resistor in line either if the coil driver is strong enough.

[crasbe]

The capacitor is a suppression capacitor, most likely for electromagnetic complicance.

[jrbe]

Sorry, I was stuck thinking points and condenser.

Yes, there is a capacitor on the +12v line to the coils.

And as crasbe said, to help eliminate emi / electrical noise. It's not required for testing unless you're having noise issues.

The second coil is wired the same as the first, just with it's own trigger.

This is how the coil fires. I think you'll have to click on the image to see the animated gif.

[Woodfie]

Thanks for the assistance.

With 12v applied , the White /green "signal" wire
is at 12v, which fits with the Typical Ignition Diagram
When the ground switch is not grounding.
Just treading carefully as have no spares coils,will try some resistance to ground to sink the signal wires 12v , to get a better idea , as could be float if it is meant to be connected to a digital 5v system...

[Woodfie]

Is C5 connected across the Igbt,?
Is the value of C5 stated?

[jrbe]

15 is Bosch speak for switched +12v.
31 is Bosch speak for ground.

No, the capacitor is on the +12v line that feeds the coils. It's value is not defined in the wiring diagram. Just dug this up,

The coil symbol clearly shows just the coils, no drivers or electronics in them.

[Woodfie]

Yes that's what I am seeing, with a 12v light bulb 50w as current limit in the white / green wire to ground, the coil draws 3.5A, with 10v still measured at the multi connector plug .
With momentary pulsing to ground, a resemblance of a spark is generated at the plug.. with this crude switching, a condenser should help, but have at least seen a spark..

[jrbe]

A normal ignition coil dwell time is about 3ms. That should be all that's needed for a good spark.

The spark energy will want to return to the brown wire / position 2 on the coil connector. Make sure there is a connection back to the plug body or engine while testing, not just position 2 of the coil connected to the battery - and the plug laying on a floating ground engine.

[Woodfie]

Thanks,
yes had made sure the plug is firmly secured to the engine ground, to eliminate any high voltage going where it shouldn't.
In past, on a prius project, had used some coil driver kits from silicon chip magazine from Jaycar electronics.. these now are long gone, but have sourced the specific coil drive transistor pair , (BU941P ) from ebay . Unfortunately a long delivery time, so may borrow from the prius set up, till they arrive.
Hope that having the brown ground wire separate compared to the circuit diagram will still work,

[jrbe]

Hope that having the brown ground wire separate compared to the circuit diagram will still work,

The capacitor on an ignition coil harnesses is typically a plastic cube with a ground / mounting tab that grounds near a coil. It's likely easier, safer, cleaner, cheaper to add the nearest coil or coils ground wires than to have a bunch of loose ground ring terminals to connect.

The engine should still be connected to the 12v system - / ground.

It will very likely run without the capacitor unless you have sensitive electronics running the engine that are getting interference from the coil noise.

You could hunt down a coil driver from a recycling yard too.

[Woodfie]

A test run with a Lucas condenser off a Bedford, across the " point " break, and a juicy reliable spark is there.
Confirming this coil over is basically a standard form of coil, with no Smarts inside..