The Turbo 3500 F23 swap (Page 68/79)
ericjon262 JAN 20, 01:02 AM
Haters will say it's photoshopped... I drove it to work yesterday, probably should have done alot more tuning first, but it made the trip ok, and I datalogged the whole thing so I could make changes before going home.




Tuning changes made while I was at work made a massive difference, but some of that may be placebo, I capped the throttle at 30% but when I did that, I put it in the 100% accelerator position row, meaning I would have to mat the gas, to get to 30%, while I was at work, I realized this was a terrible idea, because every time I raised the cap, I would have to relearn the throttle, so now, the table looks like this:



I also started looking at acceleration enrichment a little bit closer, previously, AE was based on throttle position only, with the option for MAP based AE. I decided to enable the MAP based AE, I think it is probably a good idea for a five speed car, as the clutch can be manipulated independent of throttle, and cause the engine to require more fuel enrichment. right now, I have the AE split 50/50 TP/MAP, I'll do some playing around with it and see what happens.

I did not drive it to work today, weather forecast calls for rain, my wipers work, but my headlights are terrible at best, I ordered a set of relays from Ballenger Motorsports, and am going to add them into the headlight circuit to get them a more direct path back to the alternator. I have a spare headlight harness that I will rework so I can still drive the car when the weather permits. Honestly, I don't really want to drive it in the rain even with working headlights if I can avoid it. The gear on the driver's side motor is also stripped, so I ordered a pair of Rodney Dickman's headlight gears to replace them, I'll rework both motors when I replace the harness.

I'm having issues with noise on the VSS still:



it's become annoying enough for me to begin looking at modifying a hall effect type sensor to replace the VR sensor on existing currently. I'll get dimensions of the existing sensor, and then see if I can either make an adapter for another sensor, or adapt a generic threaded sensor like this:

https://www.diyautotune.com...ffect-threaded-body/

I really don't like that one, because there's no connector, and it requires a pull up resistor, it's also not an off the shelf part that I can get anywhere, which could make future replacement difficult, it also has a threaded body, which will have to be sealed somehow.

I think a better option may be to adapt a camshaft position sensor for a LX9, I'm fairly certain it's a smaller diameter than the existing VSS, meaning I should be able to make a bushing for it to fit the F23 five speed, then it just needs to be tabbed for the hold down(s).

I recently discovered that Wilwood manufactures a drum in hat parking brake designed to fit under several of their hats. I've been considering developing a new set of knuckles for the car to accommodate a better(lighter) brake package, so I may shoot them an email and see if they have any useful drawings they I can use for my development. I suspect they will decline, but who knows.

https://www.wilwood.com/Har...?itemno=249-12660/61

------------------
"I am not what you so glibly call to be a civilized man. I have broken with society for reasons which I alone am able to appreciate. I am therefore not subject to it's stupid laws, and I ask you to never allude to them in my presence again."

cognita semper

Skybax JAN 20, 12:48 PM
Hats off to ya, keep p the good work!
zkhennings JAN 20, 03:34 PM
Does the LX9 cam position sensor require a pullup resistor? That would be a slick upgrade. I was not planning to feed VSS to my MS2 but you never know. I imagine you could just machine the stock VSS and outfit with a CNC trigger and bracket for cam sensor, would it need missing teeth or would a single tooth work with what the MS3 wants to see? Nice job on getting it driving again after some pretty major surgery.
pmbrunelle JAN 20, 04:59 PM
Everyone you work with has a pickup truck?

********************************************************************************

The M12 threaded sensor is an off-the-shelf item; DIYAutoTune didn't create anything. They're readily available on Digikey/Mouser/Newark... just not O'Reilly's, Autozone (whatever American auto parts retailers I can think of).

It's a Cherry (now ZF) sensor. I have a work colleague who swears by them. Some of them only have a 105°C temperature rating, so watch out for that.

If your chosen solution requires a pullup resistor, there's a good chance that your MS3Pro has one on certain inputs.

On my V3.57+MS3X combo, the Tableswitch In input has a pullup to +12 VDC. It's meant for use with grounded-switch inputs. I use it on my car with the Fiero VSS buffer board, which has an open-collector output for the ECU.

********************************************************************************

A VR sensor and its associated detection circuit is "one system".

Not that VR is bad, but things get complicated when you take a working system, chop it into two pieces, and then attempt to graft another detection circuit onto it.

I had troubles of this nature with my Fiero's crankshaft sensor...

With Hall sensors, the detection circuit is physically bundled together with the Hall element, so the working system stays together, and doesn't get chopped up.

********************************************************************************


quote
Originally posted by zkhennings:
I was not planning to feed VSS to my MS2 but you never know.



When I had an MS2 on my previous stock 2.8 Fiero, I did. I ran the output from the Fiero buffer board to a spare digital input, with a pullup resistor added in the proto area. I didn't have an actual speed reading; I just had a 0/1 stopped/moving flag.

With customized firmware, I gained the following benefits:

1.
Only attempt to control idle speed when the car is stopped.
If the car is in-gear, then it doesn't make sense for the computer to attempt to control RPM. If the car is stopped, then the transmission is either in neutral, or the clutch is disengaged; in either of these cases it is OK for idle speed control to work.
Stock MS2/Extra idle control does not consider VSS input.

2.
For dragstrip launches, I had a special rev limiter setting for VSS==0. So I would simply floor the gas pedal, and then dump the clutch. Once the tires started spinning (i.e. VSS no longer zero), then the rev limiter would switch to its normal setting.

Stock MS (MS2 and MS3) implementations of launch control are really goofy IMO (not that other functions don't also have room for improvement); they need the driver to push a button, in a certain order before/after pressing the clutch pedal. Cars should be easy to drive; someone else should be able to work the launch control without me having me tell them the Nintendo cheat sequence beforehand.

Operating the 2-step with the VSS worked well with the stock 2.8. Eventually, I will rewrite the launch control for my MS3, and I will see if launch control with VSS still works well with a modified engine. However, finding new ways to beat on my powertrain is not currently a high priority.
ericjon262 JAN 20, 08:01 PM

quote
Originally posted by Skybax:

Hats off to ya, keep p the good work!



Thanks!


quote
Originally posted by zkhennings:

Does the LX9 cam position sensor require a pullup resistor? That would be a slick upgrade. I was not planning to feed VSS to my MS2 but you never know. I imagine you could just machine the stock VSS and outfit with a CNC trigger and bracket for cam sensor, would it need missing teeth or would a single tooth work with what the MS3 wants to see? Nice job on getting it driving again after some pretty major surgery.



Thanks, it's been a mountain of work, with a ton more still to go, I'm waiting on a bunch of parts to get here to fix the headlights and a few other things, once that's done I'll be more comfortable driving it in less than perfect conditions.

It does not require a pullup, it was one of the reasons why I would prefer to use it. it's also a smaller diameter than the F23 VSS, I plan to make a bushing to fit it where the F23 sensor is. because vehicle speed doesn't need timing or positioning, it doesn't need missing teeth, just something teeth to count.


quote
Originally posted by pmbrunelle:

Everyone you work with has a pickup truck?



Pickup's are very common down south, that being said, most of the trucks tend to park in the back of the parking lot, I parked in the back because it had a slight grade away from the plant, and other vehicles, my car doesn't have a parking brake.


quote
Originally posted by pmbrunelle:
The M12 threaded sensor is an off-the-shelf item; DIYAutoTune didn't create anything. They're readily available on Digikey/Mouser/Newark... just not O'Reilly's, Autozone (whatever American auto parts retailers I can think of).

It's a Cherry (now ZF) sensor. I have a work colleague who swears by them. Some of them only have a 105°C temperature rating, so watch out for that.

If your chosen solution requires a pullup resistor, there's a good chance that your MS3Pro has one on certain inputs.

On my V3.57+MS3X combo, the Tableswitch In input has a pullup to +12 VDC. It's meant for use with grounded-switch inputs. I use it on my car with the Fiero VSS buffer board, which has an open-collector output for the ECU.



while I don't disagree with the above statement, the M12 threaded sensor has a couple other pitfalls compared to using the LX9 cam sensor, as mentioned above, the LX9 sensor doesn't require a pullup, the LX9 sensor has a integrated connector that I can quickly and easily disconnect the sensor from the harness, and auto parts stores are open more, and easier to find than industrial parts houses for things like this.


quote
Originally posted by pmbrunelle:
A VR sensor and its associated detection circuit is "one system".

Not that VR is bad, but things get complicated when you take a working system, chop it into two pieces, and then attempt to graft another detection circuit onto it.

I had troubles of this nature with my Fiero's crankshaft sensor...

With Hall sensors, the detection circuit is physically bundled together with the Hall element, so the working system stays together, and doesn't get chopped up.



from what I understand, due to the nature of how the VR sensors work, they are for more susceptible to interference than hall effect.

------------------
"I am not what you so glibly call to be a civilized man. I have broken with society for reasons which I alone am able to appreciate. I am therefore not subject to it's stupid laws, and I ask you to never allude to them in my presence again."

cognita semper

[This message has been edited by ericjon262 (edited 01-20-2022).]

zkhennings JAN 21, 11:05 AM

quote
Originally posted by pmbrunelle:

When I had an MS2 on my previous stock 2.8 Fiero, I did. I ran the output from the Fiero buffer board to a spare digital input, with a pullup resistor added in the proto area. I didn't have an actual speed reading; I just had a 0/1 stopped/moving flag.

With customized firmware, I gained the following benefits:

1.
Only attempt to control idle speed when the car is stopped.
If the car is in-gear, then it doesn't make sense for the computer to attempt to control RPM. If the car is stopped, then the transmission is either in neutral, or the clutch is disengaged; in either of these cases it is OK for idle speed control to work.
Stock MS2/Extra idle control does not consider VSS input.

2.
For dragstrip launches, I had a special rev limiter setting for VSS==0. So I would simply floor the gas pedal, and then dump the clutch. Once the tires started spinning (i.e. VSS no longer zero), then the rev limiter would switch to its normal setting.

Stock MS (MS2 and MS3) implementations of launch control are really goofy IMO (not that other functions don't also have room for improvement); they need the driver to push a button, in a certain order before/after pressing the clutch pedal. Cars should be easy to drive; someone else should be able to work the launch control without me having me tell them the Nintendo cheat sequence beforehand.

Operating the 2-step with the VSS worked well with the stock 2.8. Eventually, I will rewrite the launch control for my MS3, and I will see if launch control with VSS still works well with a modified engine. However, finding new ways to beat on my powertrain is not currently a high priority.



Good to know! I will probably ask you more questions when I get to this point, those are all good reasons to let the MS2 know that the car is moving.

And Eric that's cool, is the LX9 cam sensor a weird form factor like the LZ9 one? I know that MS likes GM sensors, but Subaru cam sensors have a really good form factor for your application. The URL tool seems broken but if you search for a Subaru cam sensor I think they are the same part from the 90s until 2010 so there are a ton of them. They are small, short, and have a single bolt mounting solution. Not sure if it would require a pullup resistor or anything like that. They are cheap and readily available. My LZ9 cam sensor is massive compared to it. Might be a non-issue but just a possible option.

[This message has been edited by zkhennings (edited 01-21-2022).]

ericjon262 JAN 22, 10:39 PM

quote
Originally posted by zkhennings:


Good to know! I will probably ask you more questions when I get to this point, those are all good reasons to let the MS2 know that the car is moving.

And Eric that's cool, is the LX9 cam sensor a weird form factor like the LZ9 one? I know that MS likes GM sensors, but Subaru cam sensors have a really good form factor for your application. The URL tool seems broken but if you search for a Subaru cam sensor I think they are the same part from the 90s until 2010 so there are a ton of them. They are small, short, and have a single bolt mounting solution. Not sure if it would require a pullup resistor or anything like that. They are cheap and readily available. My LZ9 cam sensor is massive compared to it. Might be a non-issue but just a possible option.




google shows a 2 wire sensor for the subaru cam position sensor, so it is most likely a VR sensor. I would be surprised is the LZ9 and LX9 cam position sensors aren't directly compatible.

I made a sensor mount, it's bigger than it needs to be, and has way more available adjustment than needed. I didn't have a good way to measure the angle that the mount would need to be to bolt to the trans, so a slot provided enough adjustment, unfortunately, I made the slot a bit too long, and it hits a boss on one of my other F23's I have, maybe the same boss doesn't exist on the trans in my car? worst case, I take a grinder to it in that area and have an open ended slot.








I wired up a connector with some Tefzel sheilded wire for the signal wire, I plan on installing it later this week, maybe tomorrow?

I also got in new gears for the headlights,




------------------
"I am not what you so glibly call to be a civilized man. I have broken with society for reasons which I alone am able to appreciate. I am therefore not subject to it's stupid laws, and I ask you to never allude to them in my presence again."

cognita semper

[This message has been edited by ericjon262 (edited 01-22-2022).]

ericjon262 FEB 20, 10:09 PM
I installed the cam position sensor in the transmission, at first, it didn't work at all, I pulled the whole thing back out, and found that I machined the adapter a little too tall. I trimmed it down, and it appears to be working, it also seems to be running much more consistently. before installing it, I was having a bad hunting idle, and it was very inconsistent. at the start of a drive, it would idle great, towards the end, it would idle like crap. now, if it runs good at the start, and I don't change anything, it still idles good at the end. now my biggest issue is sync loss at about 5400 RPM, I've hit it a couple of times, and am hoping some adjustments to my ignition coil mount will make it go away. if not I have a few more tricks up my sleeve.

I picked up a CNC Plasma cutter, and made the first cut last night, I'm hoping I should be able to make some cool stuff with it.

------------------
"I am not what you so glibly call to be a civilized man. I have broken with society for reasons which I alone am able to appreciate. I am therefore not subject to it's stupid laws, and I ask you to never allude to them in my presence again."

cognita semper

ericjon262 MAR 09, 10:53 PM
for those of you running an MS3, if your crank position sensor is in the crankcase, and not external, and you're experiencing sync loss at 3800-5800 RPM, your sensor may have particulate on it, affecting the accuracy of the sensor.





the material on this sensor was ferrous, but the sensor being removed from a 2005 model engine, could have accumulated all of that over about 18 years of service, and an untold number of miles, for now, I'm going to tell myself that the material is wear products from the stock cam, lifters, timing set, and rockers, all of which are no longer in the chat.

after replacing the sensor, the engine now revs happily to the limiter. at this point, I have restored my timing and fuel tables back to individual tables, instead of having 2 tables each (4 total) to improve resolution. while doing this reduces resolution, it makes tuning a little easier.

------------------
"I am not what you so glibly call to be a civilized man. I have broken with society for reasons which I alone am able to appreciate. I am therefore not subject to it's stupid laws, and I ask you to never allude to them in my presence again."

cognita semper

La fiera MAR 13, 07:28 PM
I had a similar problem with the crank sensor also. It was a combination of a wrong setting on the wheel decoder and the crank sensor being swollen a bit and oily at the connector. Now the engine is happy and revs crispy clean.