T87 Swaps

[kidturbo]

Here is our new 2016 output shaft sensor live, connected, and reporting good on the TCM. You can see the min / max voltage values. It shed 5 volts from it's disconnected state. Also note it actually never drops to 0v.





Next up is our Arduino Emulator running a 40 tooth pulse pattern. It outputs range between 0-5v, and the TCM accepts this pattern fine for both Input and Output lines.



Last up we have our early model sine wave sensor connected the 9v powered VR converter. I also tested with 12v power input, and no change in output noted. First thing ya notice is we are now in the mv scale when connected to the TCM. When disconnected, it outputs about the same 0-5v range as our Arduino...





What ya think ??

 

[2004LB7]

Can you put a Schottky diode on the line so that when the voltage is below about 0.2-0.3 volts the output to the TCM will be zero?

 

[kidturbo]

Can you put a Schottky diode on the line so that when the voltage is below about 0.2-0.3 volts the output to the TCM will be zero?

That is the OEM speed sensor reading 1-3v normal. Where our Arduino and VR converter sends 0-5V outputs. But the TCM will read both the Arduino and the factory sensor fine. It doesn't accept the 250mv pulse from the VR converter when connected. That is the one getting knocked down from 5v to 250mv whenever connected to the TCM. Arduino doesn't suffer the same problem, and is accepted. Which tells me is amperage related somehow.

Think I'll wait for those SSR's to arrive and try again.

 

[kidturbo]

@2004LB7 I had to read your reply a couple times to catch the drift..
I've been playing with a stack of diodes and resistors in combinations. My first thought was lower the output voltage on the VR converter from 5 to 3.3v with resistors. However that's when I noticed the converter board only reading 250mv. My knock off PC scope won't even register the pulses it's so low. But the SnapOn captures it fine.

Think what we need to do is pull the lid off this T87 again, and run the circuits down to see what soft of isolation they are running these signals through. @Cougar281 tested this same setup last week and got the same results with the Max chip. So I feel we are on right track, just using the wrong circuitry to accomplish the task. TCM want's an +5v to 8v signal, then we shal give it one...

 

[2004LB7]

@2004LB7 I had to read your reply a couple times to catch the drift..
I've been playing with a stack of diodes and resistors in combinations. My first thought was lower the output voltage on the VR converter from 5 to 3.3v with resistors. However that's when I noticed the converter board only reading 250mv. My knock off PC scope won't even register the pulses it's so low. But the SnapOn captures it fine.

Think what we need to do is pull the lid off this T87 again, and run the circuits down to see what soft of isolation they are running these signals through. @Cougar281 tested this same setup last week and got the same results with the Max chip. So I feel we are on right track, just using the wrong circuitry to accomplish the task. TCM want's an +5v to 8v signal, then we shal give it one...

Looks like a good time to throw one of those cheap Amazon mosfet modules inline fed by the VR board to boost the output current and voltage

For the diode, which after reading the post again I think I may have misunderstood what you were saying. But you mentioned the voltage on the waveform never going to zero. I read it as that was an issue and suggested a diode to push it down. But I think after reading it again that it's not really necessary

 

[kidturbo]

Yeah the one that doesn't go down to zero is the first 2 pics, the 2016 factory sensor the T87 calls for..

What we also noted is the TCM seems to look for a specific resistance value on the 9vRef line. Won't clear the circuitry related codes when I spoof it with the arduino.. So bet we put some resistance on a 9v SSR circuit and it be

Sent from my SM-S901U using Tapatalk

 

[2004LB7]

Run a variable resistor across the lines until the TCM is happy then then measure the value and put a fixed resistor in it's place

 

[Cougar281]

Run a variable resistor across the lines until the TCM is happy then then measure the value and put a fixed resistor in it's place

Good idea.. But the only possible thorn in the side is supposedly, these 2-wire hall effect sensors have some sort of current aspect, not just voltage. So we may be able to make it 'sort of happy', enough to get a value, but it still might be ticked off.

 

[kidturbo]

I didn't have time to check the amperage on these new sensors yet, but I noted the resistance across pins. So think if we just loop the Vin to Vout through a relay, it should be well within specs.

I smoked the amp fuse in my volt meter a while back. Will try to get something setup to measure what these speed signals are drawing. But one wouldn't think it could be much for a simple pulse signal..

Sent from my SM-S901U using Tapatalk

 

[kidturbo]

Here is the final of T14 connector prototype. Should have one Monday to verify the fit..

What ya think??

Sent from my SM-S901U using Tapatalk

 

[kidturbo]

I CAD..

 

[kidturbo]

Received the two new switching options today from the Chinese rain forest. Along with a package of micr-clips and some Long needed jumper wires.

Now we can crack open the TCM again and get technical on its GPIO functions. It might not be pretty, but I'll bet I can get it all wired up to receive as shaft signal now..

Sent from my SM-S901U using Tapatalk

 

[kidturbo]

Also received the T14 adapter prototype yesterday. Not bad for the first attempt.

Terminals fit and release correctly. Robin the seal and wire guide dimensions from the OEM connector also works great. That orange is actually a printed silicone seal.

However he made one mistake on the outside housing "last picture" which makes the terminal area to shallow, and the latching lever won't lock. So gotta add a few mm in hight and it think it's usable design.

 

[kidturbo]

Tested the off shelf mosfets above, and found the blue boards only switched ground side, which does no good for controlling our speed sensor signals. The green boards are P mosfets, and switch the + voltage line. The TCM accepted the square wave pulse from these boards. Playing with some resistance, I was able to produce a clean input starting with a 12v supply, and reducing it down to about 5v. Then as I turned up the RPM's, noticed the opto-isolator on these boards couldn't keep up over 1000rpm.. But it worked.

Next step was bypass the opto and try again. Which I did, and that seemed to have resolved the Hz issue. Then smoke that mosfet and called it a day. Ordered some more chips, and now am at least fairly confident on the circuitry to make this sensor conversion part work. Now it's down to stuffing some boards into a project box and find a running truck. Also prototye 2 of the T14 connector should be here in couple days. Let's hope this one fits.

Been bouncing between 3 major projects aloing with other commitements. So havent had time to explored other hardware options any further.

 

[pl_silverado]

I'll be heading to Oklahoma sometime this month, If you have a prototype ready I can swing by and we can do some testing. Just let me know if you want me to bring an 2007.5 or 2010.

 

[kidturbo]

I've been a little under the weather, but look what showed up this week.

Think he nailed it this time. Seems to fit perfectly now.

Sent from my SM-S901U using Tapatalk

 

[PureHybrid]

Is that really the only wires needed? I'd be half tempted to cut some wires and run bullet connectors lol

 

[2004LB7]

Is that really the only wires needed? I'd be half tempted to cut some wires and run bullet connectors lol

Exactly. Why run a hundred dollar connector if $2 in bullet or spade connectors can do it?

Got to be more too it then what we see here

 

[kidturbo]

Haha.. There is about 30 wires required on that 80 pin connector. I haven't sat down and wired one up completely yet. Been waiting on that adapter kit to arrive

Sent from my SM-S901U using Tapatalk

 

[2004LB7]

Well that makes more sense