Dual VR V2.1

http://jbperf.com/dual_VR/index.html
Woznaldo
Posts: 12
Joined: Sat Jul 08, 2017 6:44 am

Re: Dual VR V2.1

Post by Woznaldo » Mon Jul 15, 2019 10:56 am

Hopefully one final question. My V2.2 board previously ran EDIS, most of the mods were removed, but a jumper between the bottom of R10 and the banded side of D9 still remains. There is also the 1K resistor in place of D8. Does any of this need to be removed for either the Dual VR Conditioner of the 4 Channel Inj/IGN boards?

jbelanger
Posts: 3860
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada
Contact:

Re: Dual VR V2.1

Post by jbelanger » Mon Jul 15, 2019 5:57 pm

That's on the input side so that has no impact on the 4 channel board. And for the dual VR board, this whole circuit should be bypassed. You should remove C11 and R11 and connect the output of the dual VR board (OUT1 or 2) to a pad that connects directly to the CPU on IRQ-1 (pin 14 on the CPU). I don't have a V2.2 board in hand so I'm working from schematics but you should be able to find which pad (on C11 or R11) goes to the CPU using a multimeter.

Jean
Image

Woznaldo
Posts: 12
Joined: Sat Jul 08, 2017 6:44 am

Re: Dual VR V2.1

Post by Woznaldo » Tue Jul 16, 2019 3:38 am

Thanks Jean, I checked the schematic, but was still a little unsure. I’ve connected directly to U1 pin 11 on the underside of the board.

I’ll throw it on my Jimstim later and check that all is well. I’ll have to check how to connect the 2nd trigger, but happy with using the IGN leds for injectors and IAC1 & 2 for ignition )2 at a time).

rk970
Posts: 2
Joined: Sun Oct 04, 2020 1:34 pm

Re: Dual VR V2.1

Post by rk970 » Sun Oct 04, 2020 1:42 pm

I would like to confirm some information.
+ - polarity on a vr sensor, Is it correct when a metallic object approaches a vr sensor the voltage goes positive and then crosses to negative when the object goes away if wired correctly.
My other question is on setting ignition input is it "falling edge" or "rising edge"
Robert

jbelanger
Posts: 3860
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada
Contact:

Re: Dual VR V2.1

Post by jbelanger » Sun Oct 04, 2020 9:48 pm

rk970 wrote:
Sun Oct 04, 2020 1:42 pm
I would like to confirm some information.
+ - polarity on a vr sensor, Is it correct when a metallic object approaches a vr sensor the voltage goes positive and then crosses to negative when the object goes away if wired correctly.
My other question is on setting ignition input is it "falling edge" or "rising edge"
Robert
The problem is that the voltage polarity from the sensor to the dual VR board is inverted on the output. Also, the polarity needed on the ignition settings depends on the input circuit used between the dual VR board and the CPU in your ECU. It also depends on which input pin you used because the firmware has been done with the assumption that the "standard" circuit is used (as per the manual).

There is also the actual sensor and trigger wheel to consider which will make one polarity better than the other. You would need to use an oscilloscope to check which polarity has the sharpest down going slope and use that on the dual VR board then set the ignition input polarity accordingly.

I wish I could be more helpful but there are so many possible variations that it is impossible to give a generic answer. If you're using a commonly used setup, your best bet is to go on the msextra forum and ask what others are using. Or if you have an oscilloscope and some time, you can follow the signal from the sensor to the CPU and see which way to go.

Jean
Image

rk970
Posts: 2
Joined: Sun Oct 04, 2020 1:34 pm

Re: Dual VR V2.1

Post by rk970 » Mon Oct 05, 2020 4:45 am

jbelanger wrote:
Sun Oct 04, 2020 9:48 pm
rk970 wrote:
Sun Oct 04, 2020 1:42 pm
I would like to confirm some information.
+ - polarity on a vr sensor, Is it correct when a metallic object approaches a vr sensor the voltage goes positive and then crosses to negative when the object goes away if wired correctly.
My other question is on setting ignition input is it "falling edge" or "rising edge"
Robert
The problem is that the voltage polarity from the sensor to the dual VR board is inverted on the output. Also, the polarity needed on the ignition settings depends on the input circuit used between the dual VR board and the CPU in your ECU. It also depends on which input pin you used because the firmware has been done with the assumption that the "standard" circuit is used (as per the manual).

There is also the actual sensor and trigger wheel to consider which will make one polarity better than the other. You would need to use an oscilloscope to check which polarity has the sharpest down going slope and use that on the dual VR board then set the ignition input polarity accordingly.

I wish I could be more helpful but there are so many possible variations that it is impossible to give a generic answer. If you're using a commonly used setup, your best bet is to go on the msextra forum and ask what others are using. Or if you have an oscilloscope and some time, you can follow the signal from the sensor to the CPU and see which way to go.

Jean
Thank you for the reply. I should of been more specific. It will be mounted in the proto area of a v3 board (with ms2). 5v and gnd are from proto. Out1 to Tsel. D14 is driving a GM ignition module to a single coil.
7.25” 60-2 wheel with a vr sensor from a 90s BMW (the pole piece dia matches the tooth Width)
I know it will likely need a 10k shunt.
Can I assume if the vr is wired to go positive as a tooth approaches then ignition input capture is falling edge? (Out1 to irq1)

jbelanger
Posts: 3860
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada
Contact:

Re: Dual VR V2.1

Post by jbelanger » Mon Oct 05, 2020 12:21 pm

rk970 wrote:
Mon Oct 05, 2020 4:45 am
Can I assume if the vr is wired to go positive as a tooth approaches then ignition input capture is falling edge? (Out1 to irq1)
Yes.
Image

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Wed Dec 30, 2020 10:12 pm

I have an internal V2.1 wired up to my 94/95 Miata's MS3 PNP's +5Vref, Sensor ground, and Digital in 2 on the option connector which was per DIYAutotune's recommendation (both +5 and Sensor ground, Digital in 2 is the default VSS input). I'm using a VSS from a 2004 Miata (a round gear-driven spinny type VR sensor located in the side of the transmission, driven by a gear), and it only will occasionally give me a VSS input. Looking at my logs, it occasionally works great but other times reads zero. When it works, it seems to be flawless and works down to 1mph, all the way to as fast as I've needed it to when it's working. Then it will suddenly stop seeing the VSS input. When it is working, I don't see anything that would indicate there's interference or noise except sometimes immediately before it stops working. It just either seems to work exactly as expected, or doesn't work at all and reads 0.

Probably the best example I have of what happens is attached in a screenshot. I can't get the entire log to attach since it's a bit large. It reads the sensor with no issues while I'm trying to tune my VE table at an airport runway and at about 193 seconds, the VSS input climbs very rapidly (while the car is actually decelerating), decelerates rapidly to show a constant speed higher than I actually was moving for a couple seconds, then totally falls away to zero and stays there for the rest of the log. I've verified a good ground and +5 at the connector, and the speed sensor seems to do what it should without being scoped (I don't have an oscilloscope - but that's probably my next step to diagnose if there's nothing obvious).

Any thoughts?

Thanks!
-Eric
Attachments
VSS1 Error.PNG
VSS1 Error.PNG (44.06 KiB) Viewed 3528 times

jbelanger
Posts: 3860
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada
Contact:

Re: Dual VR V2.1

Post by jbelanger » Thu Dec 31, 2020 6:23 pm

Eric,

Can you compute the frequency of the signal at the point where the speed goes wrong (which is what I assume happens in your picture where the speed goes way up before going back down to 0)?

But that's probably not the issue. If that goes completely down to 0 and stays there when at other times you have a correct speed on the entire speed range, that sounds more like some bad connection. Check that your power connection is solid (5V and ground) on the board and on the wiring/connector. Also check the VR connection to make sure those are also solid and that there isn't any possibility of a short between VR+ and VR-. You might need to clean the board if there is any flux residue from the solder left on the board because that can become conductive in certain situations.

Jean
Image

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Thu Dec 31, 2020 11:09 pm

Thanks!

It dies at random times it seems like. Sometimes it doesn't read at all on a fresh startup, and other times it'll stop working after a few minutes. I can't find any correlation to a speed or duration when it stops working. When I moved the car to the other side of the garage, it saw absolutely no signal. I'd tend to agree it seems like a wiring issue but I'm not sure how. I was very methodical. I'm not using a shielded wire but I did twist the pair inside the sleeving.

Immediately after shutting it off I verified non-infinite (~140ohms) between +/- on the VR and was able to see a good +5 to ground through my connector. I learned my lesson a while back on solder flux, and was careful to clean it off when I soldered the board but I'll double check everything.

I might go put it on stands and see if I can at least verify the VR sensor is giving me some consistent AC voltage. I don't have a scope but probably should buy one. I'm struggling to figure out what to buy that will do the job and not ruin my winter racecar budget though... Lol.

Thanks,
-Eric

jbelanger
Posts: 3860
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada
Contact:

Re: Dual VR V2.1

Post by jbelanger » Thu Dec 31, 2020 11:55 pm

There are so many oscilloscope choices out there that it is difficult to give an advice. It also depends if you think you'll use it a lot or you just will need it once in a while and if you think you will always have a PC available. In the latter case you could go with a USB scope which will be cheaper. But don't go so cheap that you only have a single channel; you will regret that very quickly.

You can find a lot of different reviews and opinions if you google it and there has been a few posts on the subject on the msextra forum.

Jean
Image

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Fri Jan 01, 2021 12:29 am

I'll have to look around on the msextra forums for scope advice. I did a search the other day and didn't find much, so I tried to make a post but it hasn't been approved yet. Maybe they are filtering me because it's a repeat question. More digging is required. I did quickly find that a dual channel at least is good, but the bandwidth and other specs I'm not familiar with yet. I'm guessing my signal is in the range of 10kHz in this particular case, so something 50kHz would be plenty. Since everything seems to be 1MHz or better, it's probably hard for me to go terribly wrong, especially since I'm dealing with <20v in most everything automotive I've ever wanted one for.

I did a little messing around though - I went and put the car on stands, stuck it in various gears at idle, and found that the VR sensor seems to be outputting what's expected. I just checked for AC voltage at the leads and get anywhere from 1.4ish VAC at idle in 1st, with each gear increasing a bit in between. Still no input that the MS sees right now.

The Dual VR 2.1 has 5v and a ground input that are still checking good from the MS, but the output side reads the exact same voltage as the 5v input (roughly 5.01), either on AC or DC when trying to see if it'll pretend to be an AC output with my basic multimeter. I tried swapping to VR2 and Output 2 on the Dual VR, with no change.

Since it's all I really know to do otherwise, I brought the board inside and checked the resistance across each pin combo. I got the following:
Vin--> Ground = 9.40kOhm;
Vin --> Output 1 = 4.69kOhm;
Vin --> Output 2 = 4.72kOhm;
Ground --> Output 1 = 14.16kOhm;
Ground --> Output 2 = 14.18kOhm.

After de-soldering and thoroughly cleaning the board with alcohol and gently scraping off anything that remotely looked like flux (there wasn't much if any), I get the same resistance values. So I don't think I have a flux contamination issue. I had that happen on my F250 when I replaced a fuel pump relay, and it would hit the starter any time I pushed the clutch pedal in. That was fun to diagnose!

At this point I'm guessing I need to scope the output to verify that it's working unless you see something in my very limited diagnostics that looks off otherwise.

Thank you!
-Eric

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Fri Jan 01, 2021 8:51 pm

So, on a gut feeling I swapped the input pin to the MS to Digital Frequency In 3 (it was Digital Frequency In 2 as default for VSS), and I now have a VSS signal. I wonder if something is wacky with that input 2. I'll contact the DIYAutotune guys and see if they have any ideas why that may be the case. Hopefully I don't have a bad circuit in the MS! Either way, it seems to work for now at least... We'll see if that continues after I can get the car out again for some more testing.

Thanks for the help!
-Eric

jbelanger
Posts: 3860
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada
Contact:

Re: Dual VR V2.1

Post by jbelanger » Sat Jan 02, 2021 9:47 pm

Good to see you have a working setup. But it is annoying to not be certain what the issue was. It might be that you corrected some issue through the re-wiring or you may have an unstable input on your MS; there might be a cold solder on one of the input circuit component. Let's see what the DIYAutotune guys say.

Jean
Image

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Mon Jan 11, 2021 8:21 pm

Alright!

I finally got my scope in and was able to test out the signals. I found something interesting!
I scoped my DualVR connectors, both input and output while connected to Digital Frequency in 3, where it has been working. Yellow is input, turquoise is output. Everything looks normal all the way up to around 5000rpm in 5th gear, which seems to be around 105-115mph, depending on how well-calibrated my ear is. This is a video of that run-up: https://youtu.be/ii7dosdoRb0

I then moved the input pin to Digital Frequency in 2, and repeated the experiment. Interestingly, there seems to be a modification to the square wave that coincides with the peak of the VR signal. I checked all my leads and connections several times, as well as inverted the VR input signal to the scope to make sure the scope wasn't acting up, but it was very consistently creating this modified square wave. I also ran this up to 5000rpm in 5th gear, and got this: https://youtu.be/5G0S23lf3oI Interestingly, I disconnected one of the leads from the VR signal, and the modified square wave output continued to be produced, but at a lower amplitude. I also swapped the leads to see what it looked like as an inverse of the input, and there's a photo of that as well. Somewhat unexpectedly, the speed input seemed to be recognized this entire time on Digital Frequency in 2, which had not been the case over the past few weeks.

After these, I put the speed input back on Digital Frequency in 3, and was able to repeat the normal looking square wave without any anomalies. I also did this with the VR signal reversed into the scope to see if it changed anything, but it behaved as would be expected (photo also attached).

For some reason I can't get the aligned "Digital In 3" to upload. I made another post with the Digital in 3 photos attached. This post just has the Digital in 2 photos.

Thanks again!
-Eric
Attachments
Digital In 2.jpg
Digital In 2.jpg (993.09 KiB) Viewed 3275 times
Digital In 2 Inverse.jpg
Digital In 2 Inverse.jpg (1.03 MiB) Viewed 3275 times
Last edited by rccardude04 on Mon Jan 11, 2021 8:22 pm, edited 2 times in total.

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Mon Jan 11, 2021 8:22 pm

These are how it looks when working properly (Digital Frequency In 3).
Attachments
Digital In 3.jpg
Digital In 3.jpg (1.01 MiB) Viewed 3275 times
Digital In 3 Inverse.jpg
Digital In 3 Inverse.jpg (883.66 KiB) Viewed 3275 times

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Tue Jan 12, 2021 10:36 am

I got a quick reply from DIYAutotune! Initially they told me Digital Frequency in 2 and 3 were identical on my MSPNP Pro, but another guy has told me that #3 has a 100k-ohm pullup to 12v, so not quite identical it turns out.
Those are good scope shots. The phenomenon appears to be bleed thru from the VR conditioner, like perhaps there is some stray solder flux with a short between VR- and the conditioned output. Another likely cause of bleed thru is poor grounding.

Lastly, digital Freq In 2 does not have an internal pullup. Digital Freq In 3 has a 100k 12v pullup. So there is a slight difference in the ECU's internal circuitry, but not much of a difference. Adding a bit more pullup may be required to get the circuit to work with Digital Freq In 2. I found this pic of your conditioner PCB: dual_VR_v2_1_size.jpg (319×168) (jbperf.com)

It would appear that R6 and R3 are the pullups. Perhaps the value is a little bit too high, not allowing sufficient current to trigger DFI 2.
Any thoughts on this? I replied to assure him that I have the board cleaned off and that I unsoldered and installed new pins after inspecting, then did my normal alcohol clean plus scraping the board a bit to ensure there was NOTHING left for flux on anywhere I can access. Naturally, it looks a little ugly but it works the same and I know there's no flux. :)

Thanks!
-Eric
Attachments
Dual VR 2_1.jpg
Dual VR 2_1.jpg (348.03 KiB) Viewed 3207 times

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Wed Jan 13, 2021 10:50 am

Well, I scoped it with a 1k-ohm pull-up on +5 to signal output and it damped the little bump in the output, but not a lot. Also verified the ground and did some other messing around. For some reason, the input 2 started working as expected on my last swapping of wires for the night, but interestingly whenever I scope input 2, I get a 5v peak to peak wave. Input 3 sees about an 8.4v peak to peak wave. I have no idea why. Very strange!

-Eric

jbelanger
Posts: 3860
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada
Contact:

Re: Dual VR V2.1

Post by jbelanger » Wed Jan 13, 2021 2:06 pm

You either have wiring problem or a measuring problem. The output you're seeing is not possible from the board: you either have 5V or 0V at the output. If you're seeing something else (the bump in your trace or, even worse, 8.4V peak), you are either not measuring the actual output of the board possibly due to wiring that backfeeds the input signal to the output or your scope probes are not connected correctly.

If you actually have 8.4V at the output, you are about to fry the chip on the board. But due to the other issue, I would guess that your scope probes are not connected correctly and that you don't actually have a correct ground reference. Please post pictures of how the probes are connected.

Jean
Image

rccardude04
Posts: 16
Joined: Tue Dec 29, 2020 7:07 pm

Re: Dual VR V2.1

Post by rccardude04 » Wed Jan 13, 2021 2:14 pm

This was last night:

A is to the board's output signal/ground
B is to the VR's +/- input

The wires probing are pushed into the back of those little JST-XH connectors, which are typically used as a balance lead on RC airplane lithium batteries.
DFI3 Final After 4 Swaps No Pullup.jpg
DFI3 Final After 4 Swaps No Pullup.jpg (585.74 KiB) Viewed 3131 times
I also probed the ECU/board's ground to the chassis ground where the board is mounted in its plastic box to see if there was a grounding issue. It gave me this:
Ground Noise 20mV scale.jpg
Ground Noise 20mV scale.jpg (502.75 KiB) Viewed 3130 times
Changing nothing but the pin on the ECU, I got this at the end of the night before I finally went inside. Something is intermittent, it seems. This looks like what I'd expect, including a 5v peak to peak.
DFI2 and VR Final Check No Pullup.jpg
DFI2 and VR Final Check No Pullup.jpg (434.25 KiB) Viewed 3129 times

Post Reply