MS3 Pro & Cross talk from VR Sensors

[prof315]

I have had perpetual issues with 24 /1 style distributors especially the Tec/Denso ones with the sloped single tooth. I found a solution that seems to work pretty well however. Cut 2 teeth from the 24 tooth wheel and set it up as a 24-2 at cam speed. This way you have full sequential sync using a single trigger. I have done this several Honda and Toyota distributors successfully even to the point of retrofitting and old Honda distributor with a 24 tooth wheel from a later model version. One thing to note however. It works best with the factory VR sensors because of the small tooth size and gap. On the retrofit set up I was using one of DIYAutotune's threaded body hall sensors and had to make it an 12-1 wheel to prevent constant sync loss as the tooth spacing was a tiny bit too narrow for the sensor.

[muythaibxr]

24 - 2 at cam speed is the same as 12 - 1 at crank speed. I'm not sure how that gives you what you need to run seq/cop.

24 - 1 at cam speed would...

Ken

[juansh2385]

24 - 2 at cam speed is the same as 12 - 1 at crank speed. I'm not sure how that gives you what you need to run seq/cop.

24 - 1 at cam speed would...

Ken

I think he is referring to cutting 2 continuous teeth because they are really close one to another and 1 sometime do not identify correctly the missing are. I have done the same in a similar setu up.

[muythaibxr]

OK, that would make more sense.

I personally don't want to do that because I believe it shouldn't be necessary. Toyota used these things in tons of different models of car and engine from the late '80s until the late '90s. I doubt they were doing anything crazy in processing those signals... It's more likely that they were doing something in their circuits to filter out the crosstalk.

Has anyone tried looking at the circuit on the toyota ECU?

Ken

[prof315]

24 - 2 at cam speed is the same as 12 - 1 at crank speed. I'm not sure how that gives you what you need to run seq/cop.

24 - 1 at cam speed would...

Ken

I think he is referring to cutting 2 continuous teeth because they are really close one to another and 1 sometime do not identify correctly the missing are. I have done the same in a similar setu up.

Exactly

[prof315]

OK, that would make more sense.

I personally don't want to do that because I believe it shouldn't be necessary. Toyota used these things in tons of different models of car and engine from the late '80s until the late '90s. I doubt they were doing anything crazy in processing those signals... It's more likely that they were doing something in their circuits to filter out the crosstalk.

Has anyone tried looking at the circuit on the toyota ECU?

Ken

I haven't looked at a Toyota ecu but I have looked at a number of Honda ecus that use similar distributor set ups and they all use an lm2904 for each wheel but I couldn't really follow the rest of each circuit

[shaodome]

OK, that would make more sense.

I personally don't want to do that because I believe it shouldn't be necessary. Toyota used these things in tons of different models of car and engine from the late '80s until the late '90s. I doubt they were doing anything crazy in processing those signals... It's more likely that they were doing something in their circuits to filter out the crosstalk.

Has anyone tried looking at the circuit on the toyota ECU?

Ken

while they used this "type" of distributor/cps for many engines the majority didn't have that slope on the cam sensor. They had distinct points for the G signal. I've taken apart plenty of them :-/

[Danbob]

Hi shaodome, I've just been investigating a 3SGTE Gen 2 CAS and it also has the 'sloped' single G-Tooth.

I plan to borrow a scope from work to look at the effect of the tooth spiral/slope on the sensor output.
I can't find an existing scope of one on the internet, have you taken one?


My thoughts were that the effect of the slope means that the wheel continually falls away from the sensor as it travels, therefore ensuring the sensor always outputs a negative voltage up until the tooth - This keeps the signal away from zero which helps to prevent the signal crossing zero due to noise and creating a false tooth.

This is the wave form that I predict the sloped tooth will create - Different to a normal symmetrical tooth shape (Stays negative between teeth)

Although I can not confirm this CAS measured in this scope does have a sloping tooth shape (found on internet on ausrotary.com forum)

It also should mean that the Cam sensor signal can use rising or falling edge as there is no 0v flat spot in the 'gap' like you'd see in a normal tooth shape.


That's everything I've come up with so far. I'd be interested to hear if you have found out anything else regarding the sloped tooth shape CAS's?
Dan

[muythaibxr]

Here is a scope shot of the output from one of these things. Notice the crosstalk between all three sensors.

IMG_1211.jpg

[Danbob]

Thanks for that, very useful! I've been google searching for hours and found next to nothing!

So my theory about the wave shape that the sloped/spiral shaped single tooth makes might be total rubbish then, the wave looks quite normal.

Where is Zero Volts on the G1 and G2 traces? is it where the yellow G1 text is? Or higher up?
Did the CAS have the sloped/spiral single tooth shown in the post at the bottom of the last page?


Yes the cross talk is very clear, hopefully small enough inputs to be able to be filtered out using the VR trigger and hysteresis Pots on the board (Assuming MS3 Pro has these?)

[shaodome]

Thanks for that, very useful! I've been google searching for hours and found next to nothing!

So my theory about the wave shape that the sloped/spiral shaped single tooth makes might be total rubbish then, the wave looks quite normal.

Where is Zero Volts on the G1 and G2 traces? is it where the yellow G1 text is? Or higher up?
Did the CAS have the sloped/spiral single tooth shown in the post at the bottom of the last page?


Yes the cross talk is very clear, hopefully small enough inputs to be able to be filtered out using the VR trigger and hysteresis Pots on the board (Assuming MS3 Pro has these?)

MS3 Pro does not have these. Unfortunately for me, we weren't able to figure it out in a timely manner and I LOST the customer who decided to go haltech platinum, which worked on the first crank

[elaw]

I don't imagine this helps much, but I've noticed something similar on my Audi when I was using VR sensors. The Audi has a weird trigger setup with two VR sensors right next to each other sensing features on the flywheel.

In this setup one sensor senses ring gear teeth, the other a single pin. In the signal for the pin sensor (which like the G1 and G2 sensors spends most of its time with the output at zero) you could always see a little of the waveform from the other sensor. I always attributed this to magnetic coupling between the sensors, and if the input circuit for the pin sensor was sensitive enough it would create havoc with triggering.

On my car I dealt with the issue by modifying the LM1815-based VR-conditioning circuit I was using to be less sensitive. But in some versions of the OE Audi ECU (not all versions for some reason I don't understand) they actually used an opamp circuit to invert the ring-gear signal and add a little of that inverted signal into the pin signal to cancel out the crosstalk.

[muythaibxr]

Thanks for that, very useful! I've been google searching for hours and found next to nothing!

So my theory about the wave shape that the sloped/spiral shaped single tooth makes might be total rubbish then, the wave looks quite normal.

Where is Zero Volts on the G1 and G2 traces? is it where the yellow G1 text is?

It's where the labels are.

Or higher up?
Did the CAS have the sloped/spiral single tooth shown in the post at the bottom of the last page?

Yes.

Yes the cross talk is very clear, hopefully small enough inputs to be able to be filtered out using the VR trigger and hysteresis Pots on the board (Assuming MS3 Pro has these?)

MS3 pro doesn't have them. I'm actually going to be switching to the ms3 pro VR conditioners sometime in the near future, so hopefully I can come up with a solution that works well.

Ken

[Danbob]

OK thanks for clarifying,

Regarding what you said earlier quoted here:

OK, that would make more sense.

I personally don't want to do that because I believe it shouldn't be necessary. Toyota used these things in tons of different models of car and engine from the late '80s until the late '90s. I doubt they were doing anything crazy in processing those signals... It's more likely that they were doing something in their circuits to filter out the crosstalk.

Has anyone tried looking at the circuit on the toyota ECU?

Ken

Maybe Toyota did not have any problems because unlike MS users Toyota were actually using both of the G sensors. I don't have the in depth knowledge to come up with how it works inside the ECU, but I think this is the biggest fundamental difference between OEM and MS setups.

Dan

[muythaibxr]

unless they had their filtering circuitry set up to have one cancel the crosstalk from the other somehow, I'm not sure how that would matter.

I think the biggest problem is the crosstalk from the Ne sensor that shows up on both G sensors.

Ken

[shaodome]

OK thanks for clarifying,

Regarding what you said earlier quoted here:

OK, that would make more sense.

I personally don't want to do that because I believe it shouldn't be necessary. Toyota used these things in tons of different models of car and engine from the late '80s until the late '90s. I doubt they were doing anything crazy in processing those signals... It's more likely that they were doing something in their circuits to filter out the crosstalk.

Has anyone tried looking at the circuit on the toyota ECU?

Ken

Maybe Toyota did not have any problems because unlike MS users Toyota were actually using both of the G sensors. I don't have the in depth knowledge to come up with how it works inside the ECU, but I think this is the biggest fundamental difference between OEM and MS setups.

Dan

On OEM ECUs the car will still run with just 1 G signal, fwiw.

[Matt Cramer]

I suspect the second G sensor is used for faster sync-up, not canceling out crosstalk.

[MBC-RACING]

Have been fighting with this for a long time now. I just cut the g2 Tower from the distribtor, but its still able to pick up noise from it. But now it dosnt start on the wrong g1/g2 signal any longer (fucking up ignition) , I just get reason 11 and thats it. I am going to find a 1gen 3sgte distributor and remove the VR sensor from it completely.

[phoenix3d]

Hi guys..

Over 2 dozen toyota cas's used so far.. MS1, MS2 & MS3 (incl Pro)

if using all three sensor inside....?...

Crank+ to NE+
Crank- to G-
Cam+ to G1+
Cam- to G-
Cam+To G2+
Cam- To G-

If using only Crank and 1 Cam pickup
wire like...

Crank+ to NE+
Crank- to G-
Cam+ to G1+
Cam- to G-

You have to wire Both Pickup input Negative input to the Common G- wire.

No need for 10k Ballast resistor across the inputs.. Not that I have ever needed them anyhow..
Just my 2 cents worth ;-)
Cheers
Gordon

[quark2501]

There are different style cas. The late 3sgte share ground almost all the way to the sensor. Older style 1 gen 3sgte had loose sensors inside and are no biggie the same for 7m og late 4age. They are all easy to make run, but the late 3s is different and can cause some problems.