Nistune 0.10 Release

If it isnt a problem or a suggestion it probably goes here.

Moderator: Matt

Locked
Matt
Site Admin
 

Posts: 8993
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Nistune 0.10 Release

Post by Matt »

What you observe regarding Prolific is typical to 2.0.0.18 version .... as already mentioned.
Wasnt using the 2.0.0.18 drivers, todays testing was using 2.0.13.130 version which was on that page you recommended but would not work on your system (you mentioned [!] yellow mark)

http://www.vems.hu/wiki/index.php?page= ... dowsDriver
Recommended Driver [Prolific Windows driver 1.3.0 (2010-7-15)], which contains the 2.0.13.130 XP driver. For more information, release notes are included in ZIP file
Tonight's testing with this version and LogWorks was a nightmare. It would continuously disconnect but that may be because of my temporary LC-1 install not having a decent enough ground. Regardless it would not reconnect without completely turning off the car and unplugging the USB serial converter from the laptop

That same page with the drivers you have posted indicated:
2.0.0.26 (2004-06-25) seems to work on at least one windows-XP, but not recommended anyway
When installing on my XP32 laptop and using 'manually' then 'have disk' and pointing to 2.0.0.26 it results with 'invalid data' so wont install. This is because the 2.0.13.130 is already installed. Tried my other XP PC and 2.0.0.26 installed fine. Will see if I can clean out the laptop of all previous driver versions and try again

From LC-1 manual I noted the following
2.3 Electrical Grounding Concerns
The electrical environment inside a car provides unique challenges, combining high voltages and currents, low-voltage signals, convoluted signal paths, and variable conditions (i.e., fans turning on and off, or starter cranking).
When using precision electronics, it is important for ALL electronics to share a common ground. Remember that “Ground” is more than just the return path for any circuit- it is also the reference against which any voltage is measured.
Since it is not always practical to ground every device to the exact same location, here are some tips on grounding:

1. The BEST grounding scheme is all grounds (i.e., ECU, Gauges, LC1 heater, LC1 system, etc.) SOLDERED into a single lug and bolted to the engine block.
2. The next best is all grounds attached to the same source, as close as possible, but on separate lugs. This is because even the corrosion between lugs can create ground offset and noise. Incidentally, this is why many ECUs have separate ground wires for injectors vs. ECU system ground- separating high voltages and low voltages reduces noise.
3. Grounding to the engine block is usually better than grounding to the frame.
4. Grounding a gauge to the radio is usually bad- ground offset can vary with volume.
5. Grounding to an ECU housing is generally not optimal- housings are strapped to the frame for shielding, but not necessarily grounded.
6. One of the WORST things to do is to ground most of your electronics to one place (i.e. the engine block), but ground one device somewhere else (i.e., the frame). Not only can this result in ground offsets, it can also create a “path of least resistance” for high currents THROUGH a low-current device. This can result in melted wires and vaporized diodes, when, for example, starter currents flow through gauges.
Disconnects can happen because of bad grounding it appears. I dont think disconnects are a Nistune specific issue is basically what I am saying. Innovate logworks does retry on disconnect so potentially we could try the same (additional user option) but I am unsure how we know when a disconnect is intentional (ie ignition off) so a mechanism to cancel similar to Innovate logworks may be required
Matt
Site Admin
 

Posts: 8993
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Nistune 0.10 Release

Post by Matt »

Okay had to properly uninstall the software for the 2.0.13.130 and then run their PL2303Remover.exe program to clear everything out. Now am able to install 2.0.0.26 without any errors. Will try this out tomorrow

2.0.13.130 interestingly is the official one on the Prolific website but after the problems I've experienced today with reconnecting (having to manually pull the converter) and also dual devices experienced this week on the bench it shows similar issues to the 2.0.0.18 version you were using

http://www.prolific.com.tw/eng/downloads.asp?ID=31
RomChip200
 

Posts: 426
Joined: Mon May 11, 2009 7:58 pm
Location: FRANCE

Re: Nistune 0.10 Release

Post by RomChip200 »

Matt wrote: When installing on my XP32 laptop and using 'manually' then 'have disk' and pointing to 2.0.0.26 it results with 'invalid data' so wont install. This is because the 2.0.13.130 is already installed. Tried my other XP PC and 2.0.0.26 installed fine. Will see if I can clean out the laptop of all previous driver versions and try again
You need to remove any driver using the original installation program by launching it again and selecting "remove".
Then, Prolific installers oblige to restart the PC. Upon this is done, you can install a new version. You need to check carefully the driverS versionS in hw manager, any discrepancy b/w the 2 cables and you will get such "invalid data" message.

2.0.0.26 version is the only version suitable at the moment, in my case. My OS is WIndows XP pro service pack 3.
No blue screen "IRQ....LESS3 since a while.
Sometimes PC doesn't see the plug or the unplug of the cable but this is more related to WIndows USB detection.
Unplug-plug it again solves it.
Before May 2011, I used 2.0.2.21 without any problem in the same configuration. I don't understand why 2.0.2.21 doesn't work anymore. Maybe the 2 cables were not connected on the same USB bridge at this time. I have 3 USB ports on my laptop Elitebook 6930p, 2 on the left sharing the same USB bridge, 1 on the right on another USB bridge.
I had a Dell D620 before (up to December 2010), and I'm pretty sure the 2 USB ports available are clearly separated.


I can insure you I don't have grounding issues. I'm a crazy man when speaking about reliable wiring. I usually use the star pattern for grounding.
RomChip200
 

Posts: 426
Joined: Mon May 11, 2009 7:58 pm
Location: FRANCE

Re: Nistune 0.10 Release

Post by RomChip200 »


PL2303X was found to be not working satisfactorily http://www.8052.com/forum/thread.phtml?thread=97853 - this has to do with the way how the '51RD2's bootloader sends out the characters (with no space between them); so also the crystal, associated level converters and cabling will influence the chances of this gadget working or not.

[added later] A similar, but PL2303HX-based cable was found to be working OK.
It appears, that there are different versions of the PL2303 with different suffixes; and also different revisions of the same suffix chip. Also, the reason of non-working of certain revisions might perhaps be different than the chip itself, so no definitive conclusion can be drawn at the moment based on "PL2303-based" alone.

http://forum.flashmagictool.com/index.php?topic=3274.0
Matt
Site Admin
 

Posts: 8993
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Nistune 0.10 Release

Post by Matt »

Interesting. I'm thought Prolific were good (based on my Aten cable) but obviously that opinion has changed

Anyway another update tonight with current fixes. I've been testing using my various wideband units and simulators doing recording, playback etc. Will continue with more LC-1 testing tomorrow. Is the RPM missing still in the 0.10.4 version?
0.10.5
Wideband interpolation fixes for VEMS and Techedge
Fix for wideband sensor lists on log playback
Innovate MTS crash on disconnect workaround
Retrieval of log auxillary titles fixed
Low AFM enrichment changed to 16 bits
Put string for sensors which become unavailable during logging
Fix for Log player reset to 0:00:00 timestamp on stop
Continue log recording on spurious disconnect and pause on user disconnect
Handling corner cases for device disconnects during log recording
I've still got some issues in regards to the following:
- AFR vs RPM graph
- File open issue (to investigate)
viewtopic.php?f=5&t=1933
- Crash issue on vqmap resync (to investigate, may be MTS crash issue as above)
viewtopic.php?f=5&t=1944
- Improve wideband averaging
viewtopic.php?f=6&t=1954
RomChip200
 

Posts: 426
Joined: Mon May 11, 2009 7:58 pm
Location: FRANCE

Re: Nistune 0.10 Release

Post by RomChip200 »

I will give more details tomorrow about all the Prolific cables I currently have and the tests I performed.
In the meanwhile, here is some tracks:


Check out Prolific FAQ website about Error Code 10. It seems like Prolific's latest drivers are not fully compatible with so called "unknown cable makers" or "no-brand cables". It also mentions about contacting the cable maker for Prolific chip validity. Prolific points out that it does not manufacturer end-product cables.

Q Why do I get an Error Code 10 (Yellow Exclamation Mark) after I installed the latest driver?
A If you are getting an Error Code 10 with a yellow exclamation mark in Device Manager after installing the latest Prolific PL-2303 driver, please contact the cable maker of your product for Prolific chip validity and support.

Prolific strongly recommend to only purchase USB-to-Serial cables from company-branded products providing technical support. It is not advisable to buy from unknown cable makers (no-brand cables) made in China. Prolific does not manufacturer any end-product cables and will not provide direct support to end-users.


http://answers.microsoft.com/en-us/wind ... 1da?page=4
Matt
Site Admin
 

Posts: 8993
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Nistune 0.10 Release

Post by Matt »

Prolific strongly recommend to only purchase USB-to-Serial cables from company-branded products providing technical support. It is not advisable to buy from unknown cable makers (no-brand cables) made in China
Saw that last night I was going to post that this morning after reading it
http://www.prolific.com.tw/eng/FAQs.asp?ID=50

Okay with 2.0.0.26 version this ran fine with consult and LC-1 on the bench together on the one hub.

However different story this morning in the vehicle. No problems whilst the vehicle was ignition on but no engine running. However once engine running and started to rev got continuous disconnects and needed to reseat the USB converter each reconnect.

I found that one time going in reverse it disconnected straight away. Next time going up the street hit about 3000rpm and then disconnected again. I went back and swapped my cable to the Aten one. No disconnects for the full time driving (going upto 6000rpm)

My conclusion - these no-name chinese cables are absolute rubbish. They do not have much noise immunity and when they do get noise they freeze up and require a replug in. Once swapping using the Aten cable with same Prolific chipset (different driver) there were no problems with the LC-1. Also had two FTDI devices hanging off my laptop (DLP A/D converter for boost and PLMS USB consult cable) with both having no troubles

This is the cable I use and recommend (UC-232A)
http://www.aten.com/USB-Converter.htm

There are other decent ones with FTDI chipsets also I'm sure too

Should also note that despite the disconnects and having logging going with three inputs, it continued logging until manually disconnecting the last input (DLP converter). Also with having all displays open (including raw data) there were no crash issues even on LC-1 disconnects in the 0.10.5 version uploaded last night so seems to be a lot more stable, but need to investigate why my class pointer is getting wiped out (workaround was to make a copy of the pointer and use this in the interim)
sssbassil
 

Posts: 49
Joined: Sat Sep 04, 2010 8:42 pm
Location: new zealand

Re: Nistune 0.10 Release

Post by sssbassil »

low afm voltage gose from o volts all the way to 327675 volts :shock:

thats some mean volts :D
Matt
Site Admin
 

Posts: 8993
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Nistune 0.10 Release

Post by Matt »

yeah I know. I had to double check the code myself to make sure it was 16 bits (which is was)

-address EA22 or 0A22 (16 bits)
-conversion formula of 'X * .005'
- range 0 ... 65535 x 5 millivolts = 327675 millivolts

Even still 0 .. 327.67 volts is the allowable range. I might just set the top limit of the range to something more reasonable like 5 volts since no one needs to use the full scale. Nistune reports mV millivolts but I could change the scaling to volts perhaps but its still going to look wierd
raddy
 

Posts: 167
Joined: Mon Mar 16, 2009 9:52 pm

Re: Nistune 0.10 Release

Post by raddy »

Hi all, just my note to issue with USB-RS232 converters. Im using for long time just and only one brand of this converter, that seems to be very stable and quality. Its ROLINE USB to serial converter, its far above of any "no name" converters. Im using it in medical device business....Price is about 30EUR here...
RomChip200
 

Posts: 426
Joined: Mon May 11, 2009 7:58 pm
Location: FRANCE

Re: Nistune 0.10 Release

Post by RomChip200 »

Matt wrote:
Prolific strongly recommend to only purchase USB-to-Serial cables from company-branded products providing technical support. It is not advisable to buy from unknown cable makers (no-brand cables) made in China
Saw that last night I was going to post that this morning after reading it
http://www.prolific.com.tw/eng/FAQs.asp?ID=50

Okay with 2.0.0.26 version this ran fine with consult and LC-1 on the bench together on the one hub.

However different story this morning in the vehicle. No problems whilst the vehicle was ignition on but no engine running. However once engine running and started to rev got continuous disconnects and needed to reseat the USB converter each reconnect.

I found that one time going in reverse it disconnected straight away. Next time going up the street hit about 3000rpm and then disconnected again. I went back and swapped my cable to the Aten one. No disconnects for the full time driving (going upto 6000rpm)

My conclusion - these no-name chinese cables are absolute rubbish. They do not have much noise immunity and when they do get noise they freeze up and require a replug in. Once swapping using the Aten cable with same Prolific chipset (different driver) there were no problems with the LC-1. Also had two FTDI devices hanging off my laptop (DLP A/D converter for boost and PLMS USB consult cable) with both having no troubles

This is the cable I use and recommend (UC-232A)
http://www.aten.com/USB-Converter.htm

There are other decent ones with FTDI chipsets also I'm sure too

Should also note that despite the disconnects and having logging going with three inputs, it continued logging until manually disconnecting the last input (DLP converter). Also with having all displays open (including raw data) there were no crash issues even on LC-1 disconnects in the 0.10.5 version uploaded last night so seems to be a lot more stable, but need to investigate why my class pointer is getting wiped out (workaround was to make a copy of the pointer and use this in the interim)
I agree some chineses cables are "probably" wrongly designed, that may be due to bad power management or shorcuts in the external components used. I didn't find any information about the internals of these chinese versions using the PL2303xx ...
But I'm still circumspect (read below)
Most of the time, I got disconnections (both Consult and Innovate) with my black Prolific cables when starting the car IF the battery voltage is a bit low (standstill car since few days).

The 3 Prolific cables types I own at home:
Cables.jpg
(236.14 KiB) Downloaded 4118 times
The #1 (from left) is a branded Prolific one using PL2303HX VID_067B PID_2303
The #2 is a chinese "poorly designed" cables. No PL2303xx visible, no SOIC chips, everything is bonded in some glue directly on the PCB. They are recognized as Prolific cables VID_067B PID_2303
The #3 (I have 2 of them and this is the version I currently use) are the same than you, Matt. They have more or less the same PCB than #1 and SOIC chips are arranged the same way (I think they share the same hardware). They use PL2303HX VID_067B PID_2303

I've never managed to get #2 working (yellow mark in hw manager) whatever the driver version is. They are fully rubbish.
I ask myself why such cables have been made ...
#1 works whatever the driver is, even with the latest version 2.0.16.166 http://www.prolific.com.tw/support/file ... _v1417.zip

But yesterday, having #1 plugged in and 2.0.16.166, I got some random disconnections-reconnections (engine off) without doing nothing (WIndows usb plug-unplug beep is very useful in this case ...), just having the laptop laying on the seat. This is a strange behaviour, maybe related to the laptop USB power management (running on battery). Then, I switched to 2.0.0.26 and AFAIR, this behaviour ended up.
#3 works fine with 2.0.0.26, I never managed to get them working with driver above (typically 2.0.16.166 or 2.0.13.130).
Maybe 2.0.2.1 worked in the past, not really sure now, or maybe, it was on my previous Dell D620.
#3 cables are in the trend of what is written on the net, they usually work with old driver versions, not the latest ones.
But again this is applicable to my Win XP spack 3, maybe in Win7, the situation is still different.

These cables are a veritable mess to test and to understand. Laptop USB hardware architecture has a role too. And the OS ...
On top of that, I'm going to test an external hub (4 to 1) just to have the best test coverage. First tests yesterday showed such self-powered hub can provide 100mA on each plug apparently. I may come into trouble with my cables .... but I need such hub because I'm going to switch to a Viliv S5 UMPC (one USB host only) ....

On the other hand, I have some FTDI cables on the way, Prolific starting to run me crazy ... keep in touch
Image

Some pictures I found out:

My #3 cables are this one except the crystal (a metal one) and the rs232 converter is a different part # ( not a Maxim one but Analog Devices, will post some pics, Matt please open yours and post some pics):
Image
Image
Image

Never buy this one (#2 bonded version, shorter green body than the branded one #1):
Image
Image

http://hackaday.com/2010/09/11/cheap-ca ... r-project/

Some interesting extracts regarding the bonded #2 version:
i’ve got one cable like this (bought it very cheap from china) lying around that didnt work for rs232. i thought it was just crap and doesnt work at all, but now i might try it with ttl level devices. nice.
...
This hack requires you to have an off-spec USB to RS232 dongle,

I think we may conclude the #2 has no decent RS232 converter inside ... I don't ever know why I spend so much time speaking about the #2 whereas I currently use #3 ... :roll:

Another bonded version I didn't know:
Image

http://forums.adafruit.com/viewtopic.php?f=8&t=9157
RomChip200
 

Posts: 426
Joined: Mon May 11, 2009 7:58 pm
Location: FRANCE

Re: Nistune 0.10 Release

Post by RomChip200 »

Aghrrr, a nasty bug in 0.10.5 regarding fast playback ">>>" in log player , all the values are concentrated in the left bottom corner (or at least in the latest cell monitored) ... normal speed playback seems ok.
Nistune 0.10.5 fastPlaybackBug.JPG
(178.82 KiB) Downloaded 4116 times
Doesn't occur in 0.10.4:
Nistune 0.10.4 fastPlaybackOK.JPG
(203.01 KiB) Downloaded 4116 times
RomChip200
 

Posts: 426
Joined: Mon May 11, 2009 7:58 pm
Location: FRANCE

Re: Nistune 0.10 Release

Post by RomChip200 »

Another bug ....

Steps:
1) open fuel map (e.g. normal one)
2) open injection multiplier K box (or injection latency or Volt latency change) and close it immediatly
3) close fuel map and re-open it and you got what is in the picture (empty drop list)

Note: Opening another fuel map (e.g. knock or gear) makes the initial fuel map retrieving its fields.
Nistune 0.10.5 emptyListBug.JPG
(186.15 KiB) Downloaded 4115 times
RomChip200
 

Posts: 426
Joined: Mon May 11, 2009 7:58 pm
Location: FRANCE

Re: Nistune 0.10 Release

Post by RomChip200 »

Is it normal to have a crash when doing these steps !?:
1) Nistune running, log player recording
2) switch ignition off ==> not trouble here, log player stops after a while asking "keep the file ..."
3) unplugging the 2 USB cables ..... unhandled exception
Nistune 0.10.5 IGNoff_USBunplug.JPG
(158.98 KiB) Downloaded 4099 times
Matt
Site Admin
 

Posts: 8993
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Nistune 0.10 Release

Post by Matt »

1) open fuel map (e.g. normal one)
2) open injection multiplier K box (or injection latency or Volt latency change) and close it immediatly
3) close fuel map and re-open it and you got what is in the picture (empty drop list)
Have seen this before. Just opening/closing several times makes it appear sometimes and not others. Probably something not initialised or timing issue. Repeated here, will investigate today
1) Nistune running, log player recording
2) switch ignition off ==> not trouble here, log player stops after a while asking "keep the file ..."
3) unplugging the 2 USB cables ..... unhandled exception
Havent noticed the crash when unplugging, will try that out also
fast playback ">>>" in log player , all the values are concentrated in the left bottom corner
There were quite a few changes to the log player internally. Probably checked everything but the the >>> operation. These updates affected fast playback
Locked