Login Register

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Fix one thing , break another - P1875
#16
(02-03-2020, 01:36 AM)thebanjoman Wrote:
(02-03-2020, 12:16 AM)fixkick Wrote: I am still STUCK at tires SIZE why leave off such a very important fact.
endless tires made, endless rims,  I can not at all guess what tires and rims are on this car ever.
and would be nice to know, for sure,  at post 1 for sure.
virtual signature 1:
1998 sport, 4wd, 4AT, 4door, 2.3L new engine + UNKNOWN TIRES AND RIMS. (1.8L ECU and 1.8L TCM stock sill there)

You are correct.

[quote pid='12563' dateline='1580652999']
1998 sport, 4wd, 4AT, 4door, 2.3L new engine + P245/75R16 Tires (1.8L ECU and 1.8L TCM stock sill there)


I went from the Stock tires to P245/75R16 which is about 13% larger.

I used the term 'reduction box' incorrectly.  This is what I used.

https://www.ihpartsamerica.com/store/SPE...ARBOX.html    Not from the same manufacturer but this is a picture of what I am using (removed now) .


With the box in place, speedo reading 70.  GPS says Im going 70.   (and of course P1875 codes. It is not immediate but eventually will throw the codes) 

Probably should address this at the speedo sensor than the transmission sensor as it's probably going to be easier to solve

If I can grab the VSS signal from the speedo with the arduino, change it and sent it to the ECU at the difference I MIGHT be able to solve this.

I guess I am also 'mixed up' in the direction I need to go with this.   13% high or 13% low to get it back so the 4AT trans VSS happy as to me. 

I have a 1998 FSM and can copy and upload the relevant pages. ( just the one line for P1875 , i think you did show this already and is UNIQUE on THIS CAR)

I have not played with an arduino but from my reading I think one could be used to make the changes needed.
[/quote]

THANKS FOR SHARING your cable box, (gear box device) !!!


its ok, and I see now the maker of the (cable box) does not show part numbers so is unknown. but you checked the speedo and for sure the needle reads 70mph hour now so the cable box is correct but not the VSS1 output, it is running slow.
both VSS must read correctly (about)


you need this electronics, (only 2 ways to do that) there are no gears on the end of VSS1, at all. it is coil sensor 2 wire and non hall sensor and is  big pain that. more later.
CPU
or Analog.

VSS1(4AT) >>> zero crossing detector >>>>to CPU,>>> some software scaling and then send out faster square wave at say 2vdc to TCM.
then analog.
way.
VSS1(4AT) >>> zero crossing detector >>>>to FVCO to (a 10% gain op.amp to a VFCO to TCM.  (hard to do lots of parts)

be best to make a frequency to voltage converter followed by a voltage to frequency converter.
because the frequency is so low, it would be cheaper as far as parts count to just hook up a zero crossing detector to a micro controller and use a lookup table to generate the sine wave from PWM.
Youd need look up table for all speeds,  from 1mph to 120MPH or at least to what is max speed on your car, IDK.  100 should do it.
the granularity of the table only needs to be inside your (10km/m window)

10kph= 6.213712mph so make the table at every 5mph.  X axis ,  20 speeds, mph 0, 5, 10 ,15 ... up. to 100
then in each cell (and array it is called) we lay down perfect vss click speeds that I DO NOT KNOW ON THIS CAR. Speedo we do but in not same.
the software would measure VSS1 speed (in HZ) and then using that speed inputted use that to index the above data array and then output  that HZ to your CPU output port
called VSS out to TCM.  2 to 5v output will work and is a perfect square wave,  and is 13% faster all the time, 1MPH up.


The hard part is getting the arduino to measure the vss1 input,  and creating a number (integer) from 1 to 20.
the array is simple
the output too, is kinda simple using the Arduno internal clock timers, sending table data to them get you the output you want.

The advice I have  on the MEASUREMENT algorithm is to  , use interrupts,    the this routine will be dead, and each VSS1 input positive going edge, from the  zero crossing detector
would fire off the  interrupt code, each edge and be counted for HZ input. timed)
then the count scaled (VSS1 I do not know its rules) and output as 20 speeds.

the zerocrossing detector is mandatory, this device, can read the very tiny 1v and weak coil voltages.  and output a prefect say 5vdc square wave.

then that prefect square wave, is sent to the CPU (arduino)





the chips that can do the analog method are here.
https://www.globalspec.com/learnmore/sem...rter_chips

A LM2917 frequency to voltage converter chip is a monolithic to voltage converter and is used in cruise control and speedometer applications

https://www.ti.com/lit/ds/symlink/lm2907-n.pdf
see this, 
this is voltage and we then run that to a 110% gain Op. amp,  gain of 1:1.1  1volt in 1.1 volt out,
then from there to a VFC (the reverse of the above..

same place 360 eng.
https://www.globalspec.com/learnmore/sem...rter_chips

this is the reverse act. V to F. the AD7740 can do this. (your design is low frequency easy)
https://www.analog.com/en/products/ad774...t-overview

https://www.analog.com/media/en/technica...AD7740.pdf

the clock pin is the only tricky part,  Fclock-in  needs be at what ever is VSS1 Hz is at 100mph. IDK this, what Id do is measure VSS1 ouput put HZ at 10mph and 10x that or times 110%
there are clock gen chips made now today that make crystals obsolete. (ever order a custom crystal in 2020, oops they went out of business most did)
you could in fact make the OP amp in the center with a POT to set (OP.AMP) gain from 1 to 20% and now works with any tires made.
or even use $10 LCD frequency counter that can monitor output and show it on the box you make.



I can only see that the 03-72LE output shaft has this (photos blury) tone wheel in the transmission tail with may be 8 vanes ? no 4,
so if true that the output of VSS would be 4 pulses per RPM of tail shaft.. and that be 5.12 times lower RPM of the rear axles, and tire circumference be speed.(distance per turn)

but also no tone wheel in had to tell this. (suzuki secret solved) but at 10mph with a scope at VSS1 pins we see that easy.

those are my ideas.  done 2 ways, , hope it makes sense , I will edit it more times later get it better.


off my gears page the 5.12 rear axles turn 744 times per mile. stock tires. USA.


P205-75-R15

744/760 (the 2 axles used does this,
85.1 inches circum stock.

the prop shaft at 60mph stock turns at. 145.3 turns per mile.  (5.12 axle gears) (stock tires)
so if the tone wheel has  4 vanes it does  581 clicks per mile.   at 60mPH  that is one mile per minute and 581 clicks a minute, and that be ~10clicks a second, (or 10 HZ)

if my math works out.   (this signals are tiny as low as 1v and slow, slow is good, easier design using cheap chips,


OK found a video on installing VSS1 tone wheel, it is 4 vanes so is 1/2 what I said above. 


ah here is the tone wheel, special thanks go to Frank Cook !!!, for OMG wow full 4AT transmission rebuilt video, 23 chapters long. this is 4wd only example

[Image: KXObJCD.jpg]
http://www.fixkick.com
Reply


Messages In This Thread
RE: Fix one thing , break another - P1875 - by fixkick - 02-03-2020, 04:02 AM

Forum Jump:


Users browsing this thread: 2 Guest(s)