tinyIOx on MS3 and LC2

http://jbperf.com/io_extender/index.html
http://jbperf.com/io_extender/tinyIOx.html for the TinyIOx

tinyIOx on MS3 and LC2

Postby sdezego » Mon Aug 06, 2018 5:37 pm

I have read posts on your forums and on MSEFI and banging my head against the wall trying to resolve.

I have MS3 that I built over 12 years ago and have been running since. Currently, on 1.51 Gold FW. I really only want to read the Innovate LC2 Chain for Can_EGO. I am using the LSU 4.9 sensor that came with.

TinyIOx came with 0.2.0. I mounted inside the case, routed CANH and CANL appropriately to JS6/JS8 and powered from MS2 Proto Vref 5v. I see conflicting reports, but no one saying you can not power from here.

Bought the cable from Innovate to connect the 4 pin Molex now used on LC2 to the 2.5mm plug via OUT on LC2. (Verified port works with LogWorks and LM programmer)

I read all of the background, and for my case, I was under the impression that I only needed to:
1.) Go into CAN EGO/GPS settings and:
-- Fetch EGO data = Innovate Format
-- CAN Id = 4
-- Table = 7
-- Offset = 76

2.) Go into Fuel -> AFR / EGO Control Settings and:
-- Change EGO 1 Port to CAN_EGO

3.) Power Cycle

I was not getting any AFR data and double checked everything. I was miffed, but found one other post with same issue and no resolution

I did not add a New CAN device to the project at First. Tried everything, then decided to go this route.

So, I added a Project Properties -> New CAN device on ADC4 (which MS recognized device), used the proper tinyiox.ini to point it to, and went from there. I changed setting to Fetch Innovate data on 19200, etc. No Avail. So then I started pushing shiny buttons and pulling levers.

At one point, I was no longer able to communicate and though I programmed something wrong and really screwed things up. Even tried Port Scanner (Tuner Studio Closed) and only found MS3..

Tried removing CAN device, etc..

Today decided to go back, and first thing was tried port installer again, and viola,. I see MS3 and TinyIO. Tried again to use CAN EGO with no avail. Again, tried to Add CAN device with proper .ini (matching Firmware 0.2.0) and when TS shut down and restarted, I saw all sorts of Conflict issues odd stuff and CAN port conflicts and MS reports it is going to disable CAN device.

So, again, I removed CAN Device, went into project, and removed anything related to CANID, TinyIOx, etc. After many various tried, I lost comm with TinyIOx again (even after complete power cycles, etc). Could not read from Port Scanner again. Is TS disabling device vie the Tune File?

I did not try to do anything with the FIRMWARE, but if I am able to connect again, should I go the route of reloading Firmware with Port Scanner? Are default port settings and such in that for TinyIOx.

Not sure if all of this corrupted my project (left remnant settings) and not sure where to go from here.

I am attaching Project, BUT NOTE, I set EGO port back to EGO from CAN_EGO, so I can run the car. Also removed, CAN Device, and associated CANID and TinyIOx files from project folders. Are there still remnant settings that are stuck in my project? When I start TS project, it still shows (TinyIOx Alpha V0.2.0...) included in my project name. I wish I would have created a project backup FIRST...

What do do and any idea why I could not get AFR data to begin with? I see there is a Bug fixed in 1.5.2 beta 2 but does not look like it is for TinyIOx

Regards,
Shawn
Attachments
Corrado_MS3_BackUP.tsproj
Set back to EGO from CAN_EGO and removed CAN Devices and cleaned up CAN device files
(1.63 MiB) Downloaded 13 times
sdezego
 
Posts: 7
Joined: Mon Aug 06, 2018 11:00 am

Re: tinyIOx on MS3 and LC2

Postby jbelanger » Tue Aug 07, 2018 12:09 pm

Your project doesn't have the TIOx information so I'm not able to see how it's configured; the MS3 settings look ok. So please add it to your project and then also make a short datalog with the TIOx realtime data enabled. That will allow me to see the TIOx settings and if there is any data going through.

A screenshot of the port scanner window would also be useful.

By the way, powering from the proto area 5V and ground is fine.

Jean
Image
jbelanger
 
Posts: 3633
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: tinyIOx on MS3 and LC2

Postby sdezego » Tue Aug 07, 2018 12:45 pm

I tried to connect via PortScanner first thing, but for some reason, I am back to no longer being able to connect. TunerStudio is closed and nothing else running, USB on COM3, etc. Port Scanner sees MS3 fine, but nothing else. This happened the other day, but when I came back the next, I was able to connect again (but did absolutely nothing except step away for a while)

I was going to add it back to project, but when I opened the project I sent you, I received an error and the Main CAN ID was set to 4?? I definitely did not change this, so not sure how it got set. Anyway, corrected project, but can not connect to TinyIOx (as I suspected since I am not able to via PortScanner).

I will need to pull the lid off and make sure it is still powered properly, but everythign else in MS is working and car is running. I suspect something got programmed in a bad way. Is there a more forceful way I can either reprogram the firmware or attempt to get it back on-line?

sorry...

Here is the error I received yesterday from MS when it was in the project (Thus why I removed it).
Attachments
MS_CANerror.JPG
MS_CANerror.JPG (65.06 KiB) Viewed 941 times
sdezego
 
Posts: 7
Joined: Mon Aug 06, 2018 11:00 am

Re: tinyIOx on MS3 and LC2

Postby jbelanger » Tue Aug 07, 2018 12:59 pm

That sounds like your MS3 is set to CAN ID 4. That would explain your problems. Check what the port scanner shows. As for fixing this, open TunerStudio, connect to your MS3, go to the project properties and make sure your main controller CAN ID is set to 0. Then go to the CAN parameters settings and make sure "My CAN ID" is set to 0. Close the windows and power cycle your MS3. Go back and check that you do have the correct CAN ID in both places. Run the port scanner to confirm.

You cannot have 2 devices on the CAN bus with the same CAN ID because there will be conflicts and the MS3 should always be set to 0. You probably inadvertently changed the MS3 CAN ID when you were adding the TIOx to the project and that's where your problems started.

Jean
Image
jbelanger
 
Posts: 3633
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: tinyIOx on MS3 and LC2

Postby sdezego » Tue Aug 07, 2018 3:00 pm

Right, when things went crazy somehow the Main CAN ID was changing after a TS self restart when adding/removing CAN devices. I verified probably 5-10 times along the way and every time I re-added TinyIOx as a CAN device, Controller was ID 0. This was some fluke, but it happened twice perhaps from me deleting the corresponding CAN/TinyIOx files from the project dir after removing CAN device? ...maybe causing TS to change the only CAN device it found (Main Controller to what the TinyIOx was)? anyway, lets neglect this for now as I know It was correct, read on

That aside I have an update. I went down during lunch and decided to try and connect, again, with no changes, just time passed. I was immediately able to connect with port scaner.

So, I took the liberty of trying something whilst I had connection. I changed firmware down to 0.1.0d. I apologize in advance if it seems I am changing things willy nilly, but I had a reason to try this..

Firmware updated fine and verified in PortScanner. Opened TS project, changed EGO -> CAN EGO restarted TS and plugged in LC2. Immediately response and started reading AFR data correctly. Unplugged serial for a sec just to make sure it was reading CAN data and not fooling me, and AFR dropped right off. So, confirmed it was reading LC2 innovate Chain data.

NOTE: In Lieu of changing too many things, I DID NOT add the CAN device back to TS. So, no CAN devices defined in project. Since data was reading, I just flew with it.

Decided to go for a drive and all was well until about 30min in. Then AFR dropped off and signal went to err cond. Stopped, closed TS and opened Port Scanner and Nill. Changed back to traditional EGO to get home.

Got home and tried to reconnect again via Port Scanner with failure.

My Guess is, something is getting shut down due to thermal condition or something (i.e. Power to TinyIOx from Proto or TinyIOx itself)??? Is that even possible? it is the only thing being powered from proto FYI, but I am not sure if that is the same vRef 5v as MS sensors (which are def unaffected during this). It makes sense that something is getting shut down thermally thus why when I just leave it for a long period of time, I am able to Port Scan back in.

Now, I should mention this is Florida, it is about 98*F in my garage which feels like 105+F with the sun blazing. MS is in the Glove Box, but I was not running AC. As mentioned, TinyIOx is mounted inside to top of MS3 Case with Small Computer Stand off studs powered by Proto Area.

No other issues.. My First thought is that after things cool down (see if I can connect again) and possibly add the CAN device into TS and see what the settings look like. But, I would like to know reason for incommunicado after a period of time.

Sorry for the long wind, but trying to be thorough and save your time.

Shawn
sdezego
 
Posts: 7
Joined: Mon Aug 06, 2018 11:00 am

Re: tinyIOx on MS3 and LC2

Postby jbelanger » Tue Aug 07, 2018 3:31 pm

Everything on the TIOx is rated to at least 185*F or more so it should not be affected by high temperature especially if the MS it self is not. And it doesn't drain a lot of current (probably less than 50mA in a setup such as yours) so it should not be an issue for the MS 5V supply which would also affect the MS anyway.

What seems to be happening here is some sort of firmware corruption. At least that's what I think your firmware change indicates: it's not that the older firmware fixed your issue but rather reloading the firmware got rid of some sort of corruption.

What I suggest, is to reload 0.2.0. If you can't get the TIOx on the port scanner, you just need to use the bootloader jumper: power off the unit, short the two 'BOOT' pads (you can use a paper clip to connect the 2 pads), power up the unit, and remove the jumper. You can then use the port scanner to load the new code (you will see that the TIOx is in bootloader mode on the port scanner window). Before you perform any firmware upgrade on the TIOx, it is a good idea to turn off any CAN polling from the MS3 so that there are no risk of conflicts.

I wonder if you have anything set wrong on your TIOx that could cause this. To change the settings on the TIOx, you don't need to add it to the MS3 project. You can create its own TunerStudio project but you cannot use the TS detection (it will not detect the devices on the CAN bus as the main controller). So create a project but click on the "..." button to point to the correct ini file then use the serial port number of the MS3 (don't use the check button because again TS does not check correctly for CAN devices) and leave the rest as per the defaults. This should give you a project for your TIOx only that communicates through the MS3.

Once you have your TIOx project, don't load any old msq but just change the few settings you need to enable the reading of the LC2. Also save this msq and post it here. Hopefully that will get you back to a working setup. But I'm not sure if you're going to see any issue come back after a while.

One thing that could cause such an issue is if you have some marginal solder joints either on the power supply or on the CAN bus. Make sure that both ends of each of the 4 wires (5V, ground, CANH, CANL) have good looking and solid joints and that the wires are supported in some way (you don't want wires that will flop around inside the case and pull on the solder joints). You may want to take good pictures of your installation and post them here so that I can have a look and see if there's anything that looks suspicious to me.

Jean
Image
jbelanger
 
Posts: 3633
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: tinyIOx on MS3 and LC2

Postby sdezego » Tue Aug 14, 2018 5:34 pm

Sorry, I haven't gotten back to this yet as things snowballed a bit. I do, however, think I may have found the culprit. I have't been able to test yet, but will soon.

All solder joints and everything looked good, however, I found a suspect at the CAN connections on Tiox. I used twisted pair for all CAN wiring and looks like the sheathing got a bit too warm at those joints. Took a small screwdriver and separated the twist there and pretty sure this is it.

IMG_32638.JPG
IMG_32638.JPG (45.96 KiB) Viewed 915 times


IMG_32638.JPG
IMG_32638.JPG (45.96 KiB) Viewed 915 times


IMG_32666.JPG
IMG_32666.JPG (68.33 KiB) Viewed 915 times
Attachments
IMG_32647.JPG
IMG_32647.JPG (40.1 KiB) Viewed 915 times
sdezego
 
Posts: 7
Joined: Mon Aug 06, 2018 11:00 am

Re: tinyIOx on MS3 and LC2

Postby jbelanger » Tue Aug 14, 2018 5:52 pm

Once you've checked your wiring, you might also want to check one more thing. I don't know what the 2.5mm stereo plug you're using looks like but make sure that it is completely seated when you plug it in the TIOx jack.

It's difficult to say from your pictures but there is some distance between the jack end face and the MS3 endplate face. This is not abnormal but it might be just enough to create a partial connection with the serial link.

Jean
Image
jbelanger
 
Posts: 3633
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: tinyIOx on MS3 and LC2

Postby sdezego » Sun Sep 02, 2018 11:53 am

Yes, the picture is deceiving, as it actually sticks out by a hair. Anyway, I finally had a chance to get everything back together (other MS3 changes) and all appears fine now. So, looks like the CAN wires were shorting intermittently due to the overheated wire insulation on the twisted pair from soldering.

I do have another question though. I have settings to output 7.44v on error condition, and decided to test what would happen if I unplug the LC2 Stereo cable.

I received unexpected results. The A/F ratio persists from what was last read by MS (which is bad in my eyes). I expected that IOx would output error condition voltage to MS.

I tried this a few times and even unplugged at an extreme (i.e a High rev fuel cut), MS started correcting based on that value when I unplugged it.

I would like to ensure that on some Wideband failure, a Voltage is sent to MS that is out of the Correction Range so that MS will just use the MAP with no EGO Correction.

Thoughts?
Thanks,
Shawn
sdezego
 
Posts: 7
Joined: Mon Aug 06, 2018 11:00 am

Re: tinyIOx on MS3 and LC2

Postby jbelanger » Sun Sep 02, 2018 2:29 pm

If the IOx receives an error code from the LC2 or if the AFR is outside the normal range the error condition AFR value is set and that's what the MS sees. If the LC2 fails or the serial link fails, the IOx just keeps the last value because it only replaces it when receiving messages from the LC2.

Most failures will not be a full controller failure so the IOx should still continue to receive messages from the WBO2 controller and thus update the AFR value sent to the MS. That will cover sensor failures and sensor control failures inside the controller. However, there is currently no support for full controller failure or serial link failure.

Jean
Image
jbelanger
 
Posts: 3633
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: tinyIOx on MS3 and LC2

Postby sdezego » Fri Sep 07, 2018 2:22 pm

There was apparently some premature celebration, but I hesitated to post in order to do some in depth T/S.

Prior, I did what you suggested and upgraded the Firmware back to V0.2 and everything seemed fine on the bench. I had created a project just for TIOx and did not use detect, and pointed to the proper ini and set settings proper. So, things looked good on the bench, communicating and thus my celebratory post above. However, the next day when I put MS3 back into the car, I could not connect to TIOx controller again, via TS OR via Port scanner (TS Closed of course). Tried everything... 20 times...

Took MS back onto bench on Stim with same results. Could not connect to TIOx via TS or port scanner. However, I noticed immediately after power, I could connect via port scanner for a few seconds then TIOx would "Go Dark" and could not connect via Port Scanner again after port refresh. Unless I did a power cycle and again, could only connect for about 2 seconds. I tried the Bootloader and it did trigger to bootloader mode (as evident on Port Scanner), but I was not able to keep TIOx online long enough to re-burn the Firmware w/o errors.

I opened up the case on the bench and during this tested voltage to TIOx which remained @5v even during these issues. So, ruled that out. Checked wires, joints and closely inspected the CAN wires again ( for the 10th time). I and convinced thay they were ever shorting to begin with even though some of teh insulation was overheated near the TIOx joints., etc. ...still No Avail.

After about 20 power cycles on bench with MS3 opened up, I was able to keep TIOx up long enough to re-burn the Firmware, but I reverted back to V0.1d (sorry, but had to try something).

As soon as Firmware Burnt successfully I shut it down. I redid those connections to TIOx anyway just for general principle. Also, I noticed a few questionable solder joints on TIOx, so I very carefully re-flowed every joint.

Also, during this, I reflashed MS3 with my 1.5.1 s19 (which is what I had). I don't know why I did this, but just did.

Put it back on the bench, created a new proj just for TIOX V0.1d and powered it up with instant success. Let it run on the bench for an hour and no issues.

Put it on the car today and set back to CAN EGO. Ran it and refined some tune for an hour in hot weather. Not a single issue.

So, there are only 3 things that could have been the root cause (and I don't even know if some are possible):

1.) TIOx V0.2 has issues with MS3 1.5.1 (V3.0 board and non-pro MS3). Maybe some CAN issue where something errors and MS3 can not longer comm with TIOx
2.) I am very hesitant to imply this, but I really thing that it was some Marginal Solder joints on TIOx. I actually had the RS232 Stereo jack pop off all 3 pads when I was just tweaking it's position with my hand. The 3 solder joint pads just split in half like they were cold joints. Solder stayed on top of pads and top of Stereo jack pads, but split off on solder to bottom of jack metal pads. So, my gut tells me that the other joints that appeared marginal under the glass were the likely culprit. ??? I found one other post where someone had the Same apparent CAN issues which were no longer present after resoldering chip next to CAN wires on TIOx.
3.) somehow, my MS3 Firmware got corrupted initially after putting TIOx in. No Other symptoms and car ran and drove perfectly on regular EGO if the Firmware on MS3 was corrupted.

I have spent an inordinate amount of time on this, so at this point I am leaving the V0.1d Firmware on it for now and for some time until I am comfortable things are completely fixed (Appears to be the case). I only plan on LC2 data for CAN EGO anyway.

Sorry for the long write up, but hopefully this sheds some light. Hopefully, issues are resolved.

Regards,
Shawn
sdezego
 
Posts: 7
Joined: Mon Aug 06, 2018 11:00 am

Re: tinyIOx on MS3 and LC2

Postby superdavemotorsports » Sun Sep 09, 2018 1:34 pm

Shawn,
"I have spent an inordinate amount of time on this"

12 years since you built your MS3, welcome back to the Automotive Performance Electronics World (APEW for short)...
We're all a glutton for punishment here.
superdavemotorsports
 
Posts: 124
Joined: Fri Aug 15, 2014 12:20 pm
Location: Washington state

Re: tinyIOx on MS3 and LC2

Postby jbelanger » Sun Sep 09, 2018 7:09 pm

First, I am surprised about the solder issues. It would have been good to have pictures but I assume you did not take any and that it's no longer visible. Where were the solder joints you reflowed? As for the stereo jack, I am quite surprised because those are easy to see and the solder is definitely well flowed into the joint and I use flux on the pads and the legs.

What I'm wondering is if it's possible the plug was push in and taken out at an angle many times. That would put a strain on the solder joints. It still shouldn't simply break all three joints and allow the jack to simply come off like that but that's the only possibility I can see.

One thing I'm not clear about is if the LC2 was still connected to the TIOx and powered up when you tried to connect to the TIOx and upgrade the firmware. That is something that should be avoided because you will have traffic on the serial port while the TIOx is also receiving data on the CAN port when in bootloader mode; that's not something that the TIOx monitor is designed to handle. Also, you should be able to use the bootloader jumper to get into bootloader mode regardless of how screwed up or corrupted the firmware could get.

And I can't really see why you would have a more stable situation with an older firmware version. I have not had reports about an issue with reading LC1/LC2 data with the v0.2.0 firmware but that doesn't mean there is none. I have had a look at the code and I don't see where that could happen. So I would need to try to reproduce your specific setup and issue. Posting your latest msqs would help. But your description doesn't point to anything that makes sense to me (not your fault but simply something very difficult to diagnose remotely and without the exact same setup here).

Jean
Image
jbelanger
 
Posts: 3633
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: tinyIOx on MS3 and LC2

Postby sdezego » Thu Sep 20, 2018 5:16 pm

Hi Jean, Thanks for the reply.

Took it out for a few more drives and tuning, and things are still working like a dream, so issues have been resolved.

To answer your questions:
[*]I reflowed every joint on every component. The ones that I thought seemed suspect were at top right corners of CPU (Stereo jack facing left) and chip next to CAN wires. I just reflowed all for GP
[*]I did not take any pictures of this sorry. This is sort of unlike me. I was frustrated and just wanted to throw a hail mary
[*]The stereo jack issue was with out a doubt cold joints. It is the only way if could have popped off the way it did. It appears to me as if the pads on the board were Tinned, then the jack placed and soldered from the top. Joints looked good on top, but when it popped off, you can see the heat did not penetrate enough through the bottom of the stereo jack's flags to flow the joints together properly. What was left on the board, and the bottom of the flags were evident of this. In fact there was no solder left on the bottom of the flags.
[*]LC2 was always disconnected during flashing.
[*]As for the Firmware, I wonder if there are a lot of people running Firmware 1.51 with TIOx? In the changes for beta 1.5.2, I see there was a CAN issue with IOx, but have no idea if this is relevant.
[*]I may try and flash v0.2.0 again just to rule that out, but right now, I am enjoying some honey without feeling the need to poke the hive LOL

I would be glad to post my recent projects files for you. Let me gather them up.

Also, as a complete side note and unrelated. If you create a new project with V0.2.0 and point TS to the .ini as described, the new Proj will have Channel 1 (Programmable On/Off Outputs) activated. Just FYI.
sdezego
 
Posts: 7
Joined: Mon Aug 06, 2018 11:00 am

Re: tinyIOx on MS3 and LC2

Postby jbelanger » Sat Sep 22, 2018 3:20 pm

I still don't understand how those solder joints on the jack could have been cold joints. As I mentioned, the solder is flowed to the pad and leg and I also use flux on both. It is definitely not just tinned pads and soldered from the top. Pictures would have been great for this and the other joints but I do understand the frustration. I wonder if there was an issue with either the solder or the iron when I built that board.

As for the new project, TS fills the fields with 'junk' if you don't load an msq or connect to the unit before going through the different settings. So anything you see there is not 'real' and there is nothing I can do about it. You can either load the default msq (available on my site) or simply connect to the TIOx.

And if you have all the functionality you need with your current setup, just keep what you have. There is no need to upgrade the firmware unless you want something not available on the older version. There were issues with some MS3 beta versions and they should have been corrected (and were as far as I've been able to see) but there might be some combination of settings that do create some unwanted and untested issues.

Jean
Image
jbelanger
 
Posts: 3633
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada


Return to I/O Extender

Who is online

Users browsing this forum: No registered users and 1 guest

cron