3tgteu on Mega squirt?

[ra65]

You guys will probably be interested in this, posted by supermuble on celica-gts.com forums. Original thread here - http://www.celica-gts.com/forums/ind...howtopic=11116

1983 & 1984 22RE with Megasquirt Ignition
2 Modules - Stock Toyota Igniter and Bosch Hall sensor Module
(Easier than internal VR and coil driver in MS)

I built a working spark system for use with an Early Toyota igniter and distributor. This setup can be used with Megasquirt (all versions). My setup uses a Bosch module to drive the coil. The Toyota igniter is simply used as a "signal" feed into the Megasquirt.

The 22R uses a VR sensor distributor. These distributors put out an alternating current, going from positive to negative. This signal cannot trigger a coil or trigger a regular Megasquirt without special modifications. To make it easier, the VR signal must be converted over into a DC signal that is either on or off. This DC signal is the same as a "Hall" signal from most Volkswagen and Chrysler distributors. You can check if you have a VR sensor by spinning a distributor manually and checking the voltage across the 2 output wires. It should register about (.2 - .5) AC volts when turned by hand.

The Toyota igniter box converts the alternating current from the distributor wires into a DC pulse that is either on or off. This pulse controls the coil. When the igniter box grounds the negative coil wire, the coil builds energy. When the igniter box shuts the pulse off (zero volts) then the coil throws a spark. The length of time that the igniter builds the spark is determined by the speed and the distance of the magnetic trigger wheel in the distributor. At a very slow idle the spark will have more dwell and the coil will have a hotter spark.

The early (NON ECU CONTROLLED) igniter module is considered a "dumb" module. This is because it cannot control the length of time that the coil gets to charge (Dwell).

Anyway, this igniter WILL work with megasquirt. Don't throw it away! I figured out how to use it to trigger megasquirt!! The igniter can be retained because it is specifically matched to the Toyota distributor. The igniter does a very good job of converting the alternating current .5 - 50.00 volts coming out of the distributor over into a useable pulsating signal that megasquirt can use.

VR sensor ignitions are prone to radio interference from electrical noise, and the VR signal can get disrupted. That is one reason to retain the stock Toyota igniter. It already has a very nice wire running from the distributor to the igniter box and the system is not susceptible to electrical interference.

The following information shows how to wire up a Megasquirt I with a Version 3.0 board using 2 modules to control spark as well as feed Megasquirt the basic RPM input. The Toyota igniter will no longer be used for driving the coil. Instead, the negative coil wire on the igniter will simply provide a line input into Megasquirt. As always, my method is only one way to do it, but it is extremely easy and cheap.

I will make a wiring diagram tonight for more information. But here is a picture of how I hooked it up. I made my own test ignition on a piece of plywood using a stock 1984 Toyota Celica Igniter. I also used a Bosch Hall sensor module.

The Bosch module that I used is designed for Hall distributors. Hall distributors send out a square DC voltage wave. Hall sensors send a +5 or +12 volt signal and a ground signal that switches on and off. This is the same type of signal that comes from the negative coil wire on the Toyota Igniter. The Bosch module is very simple. Because it uses a simple on and off switch to trigger it, you can literally trigger a Bosch module with a toggle switch. You just need a very small pulsating current. The Bosch module provides the brute power. All you need to do is provide the small signal to it. The signal can come from a toggle switch, a Hall distributor, a Toyota Igniter box, or a Megasquirt output wire.

So here is how it works. In order to avoid using Megasquirt to drive a coil, which puts more strain on Megasquirt, you can use a Bosch ignition module to provide high current to the coil. The Bosch module will absorb the extra heat and load, making the Megasquirt more reliable. The Bosch module is also a "Smart" module that will not overcharge the coil. If you get the right Bosch module, as listed below, you will have automatic dwell control. No other modules except Ford EDIS has built in Dwell control. Normally you have to "guess" at the proper Dwell. Ford EDIS and Bosch "smart" modules are the only ones that allow automatic dwell, that means less hassle when trying to configure megasquirt.

To set it all up, you disconnect your coil from your Toyota igniter. Because Megasquirt needs a regular on and off signal to trigger it, you will need to put a tiny 1000 ohm resistor between the positive wire and the negative wire where your coil used to be. Now throw away your Toyota coil. Run the negative coil wire into megasquirt on the proper ignition input pin. The tiny resistor is necessary because it keeps a tiny amount of positive voltage on the negative igniter output wire when the igniter is switched off. This makes the igniter output wave the same as a "Hall" sensor wave from a VW distributor.

The Megasquirt will read the Toyota ignition pulse like a digital wave and it can process it according to the settings you want to use. For the output, you simply send a wire from the Ignition output on Megasquirt to the Bosch Hall sensor module. The megasquirt sends a tiny low voltage pulse to the Bosch Module. When it receives the pulse, it will fire a coil.

You need to get a Bosch module and matching coil from the same car. Any module with the following part # is a smart module and it will control dwell. Because of this you don't have to worry about burning up the module or burning up your coil, or losing your megasquirt while driving down the road. Even better, unlike an HEI system, the Bosch module can receive a direct full 12 volt signal. You don't need a special relay or resistor to provide power to the module. It is very sophisticated for its size and is very rugged, it comes with a large heat sink on OEM applications. Be sure to get the heat sink and use it!

So here is a summary, if it wasn't clear:

You will use the stock "non-TCCS" igniter to trigger Megasquirt. You will use a Hall sensor Bosch module to receive the ignition output from Megasquirt. This Bosch module will also provide the correct amperage and voltage to drive a coil. It will also provide the correct dwell. The only thing the Bosch module needs is an output from Megasquirt.

First, You remove the stock Toyota coil. Jumper a 1000 ohm resistor between the positive and negative leads on the igniter where the coil used to be. Run the negative coil wire into megasquirt. See the following pictures for more information.

The reason that you would hook things up this way is very obvious. I have researched this a great deal and have found the following benefits to using external modules to run the ignition.

-The coil driver on the Megasquirt 3.0 PCB (the VB921) can burn up if you do not set the Dwell properly with the coil that you are using.

-The Bosch "smart" module can automatically set the dwell period. If used with the matched Bosch coil, you will not burn up the Bosch module or coil due an improper setting in Megasquirt.

-The Bosch module is external and it can be replaced for free with another junkyard module if it burns up. It is easier to access because it can be mounted in the engine compartment. They are waterproof and heat proof.

-The Megasquirt V 3.0 PCB board has an option for VR sensing. This VR sensing circuit is not as user friendly as a Toyota igniter. The VR wires going from the distributor to the computer box could receive radio interference from spark plug wires, or other electrical noise. The internal VR circuit in Megasquirt is adjustable via two tiny rheostats. Sometimes the VR signal won't work and these tiny adjustments must be made on the Megasquirt board to make the VR signal trigger properly. Sometimes the VR sensing circuit works improperly causing sporadic operation of the ignition system. Overall the VR input circuit on Megasquirt is slightly less user friendly and more difficult to implement than a simple existing Toyota Igniter.

-The Toyota igniter is rock solid. It is immortal.

-The Bosch modules are very reliable.

-This system makes Megasquirt far easier to implement for spark on the 1983 and 1984 22RE engines with the mechanical distributor. For more info see pics.


The following Bosch modules are "smart" modules. If it isn't listed here, it is a "dumb" module and it will not control the dwell. It could cause a problem with your coil unless you know how to set the dwell properly in Megasquirt. I recommend using only the smart modules.

BOSCH MODULE NUMBERS#

0 221 600 001
0 221 600 006
0 221 600 008
0 221 600 011
0 227 100 100
0 227 100 101
0 227 100 103
0 227 100 104
0 227 100 113
0 227 100 116
0 227 100 118
0 227 100 137
0 227 100 138
0 227 100 139
1 227 010 001
1 227 010 004
1 227 010 100
1 227 022 008





See next post for wiring diagrams.

[ra65]

And more....

Here is a diagram I made on how to wire the old (non-ECU controlled) igniter, with a Toyota VR sensor distributor (new or old) to Megasquirt. There is also a diagram to show how to hookup a Bosch module. You can see that each module does a separate task. I have updated this post to show the final project. It has been tested and it works the way I thought it would when hooked into a Megasquirt using Megasquirt and Spark - Extra running on the version 3.0 PCB with a Version I processor.

When using Megasquirt, you won't need an adjustable distributor. You will need to lock the distributor to the engine block and remove any advance mechanisms.

The newer distributors that have no advance should work great, as they also have a VR sensor. Alternately, if using the old distributor with vacuum advance, you need to simply put caps on the vacuum canister (disconnect the vacuum hoses). Then simply remove the weights and springs from the mechanical advance, and lock the magnetic trigger wheel to the shaft by welding, thread locker liquid, or bolts. The distributor may need to be modified slightly so do not make any permanent changes until after the car is started and tuned.

As far as the Megasquirt input, I used the 12 volt Hall input, High to Low signal. (per Megasquirt and Spark - Extra instructions in the manual). With the setup I've described, the signal is high (12+) all the time because of the small resistor. When a rotor blade passes the VR sensor, it triggers, thus providing a ground signal to come out of the black igniter wire. The black igniter wire will be grounded only for a brief moment and then it will go back to being positive in between each rotor blade.

HALL SENSOR ON V 3.0 PCB

For the ignition output in Megasquirt using the version 3.0 Board with MegasquirtnSpark-Extra, it is very easy. See picture below and the Megasquirt link.






Bosch Module Wiring with PCB 3.0 - Mega Manual



And here is the final completed Megasquirt with the setup that I have described. Neither the Bosch module, Toyota igniter, coil or Megasquirt board show any abnormal heating. Even with the power on continuously (non switched) there are no heat issues. (This is due to the Bosch module automatically preventing coil overcharging - excessive dwell.)

I am very happy. I've retained the stock ignitor and coil (both Toyota). My goal was to avoid using an internal VR sensor and internal coil driver in the Megasquirt case. Further testing will be done when this system is implemented on my car. The good news is that this system should work on "any" VR sensor car (including all Toyota's) from any year.

As a note, I chose to use a 12 volt Hall input signal on the board for testing. I found that a 670 ohm resistor on the coil wire gave approximately 8 volts between trigger events. The 1K resistor only gave 6 volts, so I decided on a 670 ohm resistor. I cannot determine which resistor is the best one to use at this point.

[gearb0x]

no, they run one spark on compression, the other on exhaust.... means one spark is strong and the other weak. If they are both in compression they will probably both be weak, and no one does it this way (obviously for good reason).

Considering either configuration will use the same amount of hardware, it would be interesting to see back to back tests of either configuration, if anyone could be bothered.

[thechuckster]

dual-post coils out of a waste-spark system wont work for dual spark - to get two sparks you need two coils

[MWP]

Has anyone looked into this?
http://www.vems-group.org/wiki/

It was the MegasquirtAVR project (MS using an Atmel AVR uC), but has now been renamed and moved on quite a lot.

Support for 8 inj (sequential), 8 ignitors (COP or equiv), WB02, EGT, make it look quite attractive.

[thechuckster]

i think the Euro conversion rate made it difficult for Aus folks is probably not so bad now.

Also, the MSefi forum administrators seem to consider mentioning VEMS an extreme sin - am not sure at all what caused the fallout but the split did piss a few people off.

[XR Pilot]

I'm not 100% sure, but I think the work VEMS derived from the MS group was not acknowledged, and the developments they made ontop of MS weren't given back to the MS community - they basically copied the MS for a profit.

[3jcelica]

The Witzl,

Thanks for the tip, looks like that would be the go. The haltech or the farnell ones would both require some adaption plate as you have made for the 18rg also.

The nissan tip got me thinking that the RB30 (skyline and VL) use one that looks identical I wrote these off because they looked baulky (full of switches), but it turns out they are a potentiometer based TPS also and are more readily avaliable. Will have to test my skyline tonight and let people know.

Roger

[thechuckster]

I got a TPS from a nissan-powered holden wagan - so it would have been an RB30? major PITA is the main connector on the TPS only carries IDL and WOT switches so the smaller loose connector is the one to use - and after a lot of movement, the wires in connecting the connector to the TPS look a little frayed/damaged.

[The Witzl]

sounds like what Chuck and I have used is the same animal

[thechuckster]

except it was lazy - i took the entire Throttle body and TPS

no need to adapt TPS to anything.

[3jcelica]

I went the full throttle body last time off a camira which adapted easily just the throttle cable was a pain.

Looked at the skyline last night I can see what you mean by essentially two sensors in one. So it looks as if any VL commodore ( I keep telling people commodores are useful ) or any 3L skyline can provide the TPS.

I finish the wiring tonight so should be able to test it all in the next week or so.

[thechuckster]

If you were running a drag-racing engine and wanted to trigger NOS i think the RB throttle-body might be handy - this is assuming that the switch in the main body of the TPS is a WOT switch and that were you wanted to switch the gas on.

Personally, i've found that throttle to be too big - for a 4cyl running moderate levels of boost, a smaller diameter would give you more flexibility. Because i run MAP-driven accel enrichment, it's been difficult to get a reasonable accel/throttle response across the pre-boost MAP range without it being highly or underly sensitive.

I was tempted to try a Hyundia throttle but at sub-40mm it might be too small for engines that flow a lot of air on full noise.

Now that i'm converting to 1G, i've stopped any further development.

[3jcelica]

I have the engine running yeah. I ended up using the GM HEI module. I found that the gap between the VR sensor and the vane was two great for the GM module (worked ok for the toyota module) but this is easy to fix.

I had the engine at 5500 rpm not under load at it was smooth through that rev range. This was running under limp home mode (not ecu controlled) so the engine would not push much past that. Currently only running one coil but will work on two in the future.

So from my exprience I am fairly happy with the GM HEI with VR signal conditioning and ignitor in a neat package.

Will hopefully be getting the car on the dyno before the end of the month so I will post up results when I do.

Roger

[thechuckster]

fwiw: i used the air-gap specs for a 22RE (from my Haynes manual) when setting up my dizzi - from memory, it was quite a small gap.

good to hear there's more MSefi units on the go.