[GDCobra]

Out of interest where does the value on your print-out come from?
Is it an actual value where the coil is firing or is it from the ECU and hence the value which the management system is intending to fire?

If it is the figure which the system has decided to fire the ignitoin then the issue must lie within the ECU, if it is the engine position where the ignition is occuring then it could be the crank position sensor hardware which triggers it (generator rotor) or someother piece of hardware which is at fault. I'm assuming given the name of the tool that it is querying the ECU via OBD?

 

[GDCobra]

the value is taken by the diagnostic socket from an ELM box (K-line protocol). So I think this value comes from the ECU

I don't think the ECU is bad, as my idle is perfectly stable? Maybe I'm wrong. In France it is very difficult to find an equivalent motorcycle to test another ECM. Or a used ECM

This is the hardest diagnosis I've ever seen

That's what I thought, in which case this is what the system is calculating (and presumably is what is physically happening) so why would it do this? Seems like it could be a problem with the ECU causing a bad calculation but I think that is unlikely (although I've been surprised before!) I'd expect it to either work or not. So why would it calculate a value which is unsuitable? My thoughts are.
A sensor is giving a bad reading - RPM & MAP look OK but I notice that the throttle reading is just over 22%, presumably you are idling with the throttle closed so that is a bit surprising, may or may not be a problem.
Code inside the ECU responsible for the calculatoin had been changed

Also worth considering that your reader may not be interpreting the value correctly.


Does the reading change if you operate the clutch, put it in gear or open the throttle?
Are there any error codes present?

 

[GDCobra]

That's quite right !! but I have been looking for months which or where there is an interference

My TPS is adjusted with the stroke above 2000 RPM

nothing happens when I gear up except that the diag sees it mentioning "CLUCH=1"

Here is a video I made with the same application but emulated for windows

It's difficult to tell if the TPS figure is correct or not without knowing what the software is telling us. For instance I have a simiar piece of software and it gives 2 figures for TPS one is an absolute figure the other filtered or rationalised (I can't remember the terminaoldy and don't ahve the software on this computer), basically the sensor gives a voltage output (between 0-5v) rising with rotation of the throttle shaft this would resolve into the 0-100% but in reality the figure never goes to 0v with throttle closed or 5v at WoT so the software takes a reading at closed throttle a second at WoT and this allows it to give an accurate %age reading. I hope that makes sense, I realise English is not your first language.
In the case of your software it may only be giving the raw reading, so a value above zero may be correct although a value as high as 20+ seems high.
Did you put the bike in dealer mode and check the TPS adjustment and any other EMS errors?


I guess the question here is "Is the timing figure correct/incorrect?"
I know you have a definite problem and some thing which seems to point to the cause but I don't think this has proven cause and effect just yet.
I do see that when you increase the RPM (and to a certain degree the load) the figure does change which indicates the system is doing 'something' which I'd say is good although, stragely (maybe) the figure seems to lose the negative sign at some points which would indicate firing after TDC presumably, maybe this is due to your software failing to "keep up", without knowing the software it's difficult to tell.
That really is the problem, there are too many unknowns here, we don't know what good looks like with this software.
Ideally you would put the software on another (known good) bike to compare the behaviour and get a better understanding of what the software is telling you. Even running it on a different model of bike may tell you something.
Failing that it would be useful if someone else could run the same software and show you the results. Do you know anyone else who uses it or is there any comparison video on line?

Do you know if the timing calculated by the EMS is being reproduced physically. Bear in mind the EMS only knows what it is told. If the sensors 'tell' the system it is at position which corresponds to the calculcate ignition timing value it will fire that plug, BUT if a sensor (E.G. crank positoion sensor) was out of position (Sensor or trigger wheel relationship changed) this would result in the actual timing being incorrect.
On a car engine (which I'm more familiar with) I'd simply put a strobe light onto the crank pully and check what is actually happening (Having first checked my timing marks against piston position), this takes the EMS and monitoring softwar out of the picture.

 

[GDCobra]

Hi Toto
That's interesting, I have one of those boxes around somewhere, I'll see if I can find it and if it will talk to the bike and the software, if so I'll let you know what mine looks like, will be mid-week at least though as I have to find the box and my bike is not quite running at the moment. I'll buy another ELM327 if I can't find mine or it doesn't work.

Just had a look, seems that app is only available on Android, I have Apple phone. Not a dead loss though, I have an Android tablet it may work with and it looks like it can also be used on Windows. I'll keep you posted on how I get on.

 

[GDCobra]

I've managed to find my ELM devices and get SZ Viewer loaded onto my tablet. I'll try and plug it into the bike tomorrow (need to sort the wiring) and, if I remember to get some petrol I can hopefully test it (if the ELM devices I have are compatible - If not I'll get something on order).

 

[GDCobra]

Hi @toto13
Bit of an update. I managed to get SZ Viewer working after a bit of a false start and have some data. Don't know how meaningful this will be as I'venot been able to spend too much time with it so far but at least you know progress is being made.

I have two ELM modules, tried wiring them up and got no joy, first one just didn't give any lights (has 4 LEDs on box similar to yours) second one made a popping noise and gave off a bit of smoke! Then I realised that in my haste I'd got your wiring diagram wrong and had the image of the bike's connector upside down. Doh!

Obviously one module is now fit for the bin and thinking I may have totalled the other I tried it in my car, I got it connected to my tablet (eventually) and even got some data from the car (Mecedes not Suzuki) into SZ Viewer which was quite good.

I then went back to the bike and very carefully made the connections again and this time it worked.

It did indicate that my ELM module was a "fake" (which I'm not surprised by) and may have limited functionality, I've no way of knowing if this is the case, I also don't know if all these cheap units are fakes and how you know if you're getting an non-fake.

This is the best screen grab I could get, had to take a photo of the table screen with my phone, don't recall how to screen grab directly on the tablet.

First thing I can see which is "Interesting" is my throttle reading is exactly the same as your first reading at 22.353% (I see the one in your video was slightly lower) don't know how that is possible.
My timing figures are possibly higher than expected at around -21° but not as high as yours, my bike is running a TRE via a gear position indicator but I don't think that affects neutral, I didn't get around to putting the bike in gear.

My STPS is slightly larger than yours but not massively and could just be conditions.

As you can see from the other readings my bike was fairly cold during the test which probably/hopefully accounts for my higher fuel duration and lambda reading - Hope so!

Clutch shows 4 when engaged, changes to 20 when I pull the lever.

Don't know why my screen format differs from yours, may be down to the verison of Android being used, mine looks similar to your video image, don't know why yours had Min/Max data and whether that indicates an expected range of values or the min and max reading monitored.

Unfortunately my bike is still in a few pieces as I've been servicing and making a few mod's, much as I like the smell of burning hydrocarbons it was getting a bit overpowering in my garage so I had to shut it down! I also didn't like the way the bike was running, idle was low and it didn't sound too happy, I've just modified the fuel system to an external filter, hope I didn't get anythign wrong there or forget to connect anything up although I presume I'd be seing a DTC if that was the case.

If the weather plays ball I'll get it out on to my drive tomorrow and test again getting it fully warmed up this time and spend some more time with the software.

 

[GDCobra]

Hi,
Many thanks !

I'm surprised that the SZ VIEWER worked on Mercedes, it's supposed to be a protocol for Suzuki only

Must admit I was surprised by that too, I'm sure there was not much data there but it did communicate and what was there seemed viable, although I've not looked in detail this was more of a test that the ELM module was still working after my faux pas with the wiring!

I'm very surprised at the level of your idle .. 780 rpm ??

Me too, the bike was running slow but the reading on the bike's tacho was higher than that, I am a bit worried that something may not have gone back together correctly since my changes to the fuel system, need to check all that but also anxious to see what happens when the bike gets fully warm.

For the TPS, we are both within the standard of the workshop manual, 22% is equivalent 1.10-1.14 V

Aye, but amazing that we both saw the same reading down to 3 decimal places. Where did you get that manual data by the way?

On the other hand, your consumption seems high to me at idle: 1.5 Ms.. .

Possibly, the exhaust did small rich but as I said it was in my garage and running from cold so would smell a bit. What sort of value do you get when yours is still cold?

...it is perhaps the TRE which also influences the injection advance

I don't think the TRE should affect the ignition on time, my understanding is that this should be ignition only, I also think that a different map is used when the bike is in neutral, from what I read in the instructions for my TRE the signal for neutral simply passes straight through it is the signals for the lower gears which are "tweaked" to fool the ECU into thinking its in a higher gear.

Indeed, there is no fault code. The graphics are different depending on the version of Android

There were some fault codes when I first started the system but they were listed as historical (some at least may have been if I switched on while I had some of the airbox sensors disconnected recently) I cleared them and none came back.
I suspected the graphics was due to the Android versions but I'm a little surprised that I don't see the Min/Max listings particularly as I have plenty of space on the table screen, maybe there is a setting for this or maybe my tablet is just too old.

What year is your SV 1K?

It is a late 2005 model so is the later style with black frame, Aluminium subframe which I believe has a more capable ECU (32bit instead of 16) it has 2 map sensors on the airbox.


I'm looking to see if I can get the corresponding plug for the diagnostic socket and one for the ODBII so that I can make a proper cable for this, my current setup is a bit "Heath Robinson".

 

[GDCobra]

View attachment 118314

hi @toto13, out of interest where did you get the part number for the bikes connector and is this for the connector on the bike or the part needed to connect to it?

 

[GDCobra]

I found a screenshot on an SV 650 K7 :

View attachment 118333



it confirms that the negative value of ignition advance is not a problem

Yeah, I was going to mention when I first looked at the screen just after starting the engine I noticed that the value was only about -16 but when I took the photo had changed to -20/21, I suspect this is adjusting as the engine warms.

My consumption was not bad when I last ran the bike, hopefully will not have changed as a result of my fuel system mod's but bear in mind my bike is not up to temperature in my screengrab and my injector pulse is not that far from yours, like I say I need to get the bike up to temperature before I make any judgement, I could only run it for a few min's yeseterday.

At least this seems to answer your question regarding the issue with ignition timing and you have introduced me to this application for which I thank you very much.

 

[GDCobra]

hi @toto13 and anyone else interested.

I managed to do a more thorough test this evening, slightly mixed results though, bottom line is I'm not confident in the data I'm seeing. This could be due to my ELM module, my bike's ECU or the software but in short these issues may be my problem only or general.

This photo is before staring the bike.

Throttle value is simlar to yours, Abs.pressure is presumably the reading from one of the MAP sensors hence showing the pressure in the inlet tract, with engine off this is correctly around 100kPa. However this is outdoors in the UK in January, bike has been in the garage all day (not started since yesterday) but that's not heated so the engine temp and intake temp of around 20°C is not correct, would expect 6-7 so something is off there.

Still with the engine off but holding the throttle to WoT, note max reading is just over 86%. I'm assuming this is normal and telling me I'm at 86% of the 0-5v range, would be 4.31v which is in accordance with the figures from the manual.

This is a short while later, engine is still cold as can be seen on the bike's display but check out the RPM on the bike's tach compared to the reading in SZ, it's about double. I've no other way of measuring the actual RPM but based on the sound and the fact I'm familiar with my bike's tach I'd say that is the true reading, I certainly don't think it's reading out be a factor of 2.
Note also the ignition values at -14

Slightly later still. Bike warming up note temp reading of 57°C but SZ reading of 98, RPM has dropped but SZ still reading about half of bike's. Note timing value now at -29, seems to be indicating higher advance as temperature rising.

In gear now, GIPro showing 1st (which is correct) SZ showing 6, this is indicating that the TRE is working, it has picked up the correct gear value from the selector switch but forwarded on a false value so ECU will use timing curve without limitation. All other gears show 6 in SZ - Good.

Temp in SZ now showing 122°C, dash showing only 72, bike eventually got up to 86 at which point SZ was showing around 145.


In short some of these values are definitely not correct, I'm fairly confident in the tach reading and I know the bike's temp' reading is correct withing fine limits so some data value are untrustworthy which leads me to mis-trust all of them.
The TPS readings check out both for closed and wide open throttle so I'll accept them.
Temperaturs and RPM seem to be out by a factor of around 2, I'd say probably not quite as simple as that as it's more like 1.8.
All the Fuel-1/2 readings look wrong based on others we've seen, again by about a factor of 2, I don't think the bike is running that rich.
Gear position is working well with TRE fitted, I'd like to try it without but don't really want to mess with my wiring and can't see why it would be out anyhow.

Wish I knew what the lambda or Intake air pressure-2 readings meant, maybe I'll find out eventually.

What about the Ignition values which started this thread?
On the plus side my figures agree with Toto's which would indicate that this does not indicate a problem but what I find strange is that when the throttle is blipped the timing value rises (bear in mind we are staring negative) and actually go past zero, I saw value of at least +9, don't understand why that would happen, seems strange to be firing the ignition after TDC.

I think I may try looking for an alternative ELM module and see if I get better results.

 

[GDCobra]

And that SV650 screenshot makes me concerned for its charging system...... 12.5v battery ? and its running. minimum should be 13.1v and that is borderline.....so maybe its a little of too...

Well spotted, I meant to mention that earlier, that's low even for the battery. Hope that's just a bad reading on the diagnostic.
The SZ Viewer seems like a nice piece of software, I like to have a system which I can query the ECUs of vehicles I own, but I'm struggling to trust it at the moment. As mentioned earlier it could be something related to my system rather than with the software.

 

[GDCobra]

Indeed, there is a real interpretation problem with your device GDCobra . For my part, the idle is partly consistent with what I have on the speedometer,
just like my engine temperature which is exactly like my thermometer

My water temperature is 106° maximum until the fan starts = the data is perfect

Obviously there are a number of parts involved in this and if they don't all work correclty and in unison the result can be incorrect.
It could be my ELM module, or the tablet which is quite old and, althogh it's up to date, may not take the latest Android versions and hence the latest SZ software.
I was thinking about it last night and I know that my bike tach is reliable as I made up a TB balancer which monitored the vacuum and could work out the time betweeen each pulse, I put a calculation into this to work out RPM and it agreed with the tach.

Regarding the intake pressure, you should be at - 70 kpa (Given repair manual) which corresponds to the vacuum created in the body throttle.

My readings are not massively different to this so may still be within the tolerance, this will change with throttle position and airflow and possibly other variables, my reading is showing a lower pressure (more vacuum) which I'd think is a good thing.

The other pressure sensor corresponds to altitude, i.e. 101 kpa is equivalent to sea level.

Ah, that's interesting. I know that there are 2 sensors attached to my airbox but I thought these went to each throttle body, could have sworn I'd traced the pipes back to there. I also thought this was one of the differences/improvements over the earlier bikes. Perhaps there is a third sensor for ambient pressure?
Also strange that my reading is moving around a lot, in the pictures I've attached you can see it reads anywhere between 99-166 while the engine is running and 187 when it was stopped!
I'd have thought that with the bike static and at sea level (I'm not far above that at home) this would be constant. the first picture shows a value of 100.939 for Abs.pressure with the engine stopped, this is correct for ambient atmospheric at my altitude. It's a bit worrying that this value is changing, if the the ECU is seing this signal varying I would expect this will have an effect on the fuelling.

I've just done a Google search and found a thread (on this forum) which describes the ambient pressure sensor as being located under the seat, does anyone have an exact locatoin for this and/or photo of what it looks like?

 

[GDCobra]

Looked a bit deeper into the air pressure sensors in the service manuals

Early FI wiring diagram (Presumably Pre-2006)

Note a single Intake Air Pressure sensor and an Atmospheric Pressure Sensor


Similar diagram for that later models.

Now we have 2 Intake Air Pressure sensors (front & rear) and no longer have the Atmospheric one.
Looks like I wasn't going crazy remembering my sensors were piped to the TBs but makes me wonder how I can be getting a reading in the SZ Viewer. Looks like it could be my ECU which is not compatible with SZ Viewer.

 

[GDCobra]

There are two versions of ECM on the SVs. From 2003 to 2005, and from 2005. Also on my DL K1 it's a Mitsubishi ECM, and on yours it's a DENSO ELECTRONIC

Yeah, I'm aware that the ECUs changed, I believe the later one is 32bit as opposed to 16 so presumably a more capable & modern (for 2006) unit.
I also did a bit of digging on the OBD protocols. Suzuki were using K-Line in those days and it seems there was a change in this protocol around 2004, it's possible that when they upgraded the ECU they switched protocols.

to my knowledge, there are not two AIP (Air Intake Pressure) on the two versions of SV.

I'm not sure if I'm reading that correctly but I can confirm that there are 2 MAP sensors on my bike (both fixed to the airbox) each is piped to the area of the throttle body after the butterfly so they are manifold sensors not ambient pressure sensors I also don't see an additional sensor for ambient so pretty sure there isn't on which is consistent with the later diagram.

SZ Viewer should be compatible with all Suzuki ECMs. I think there is a problem with your ELM, and not your version of android

I suspect you are probably correct, I was going to test on an alternative android unit as I thought I could get hold of a fairly recent phone but that has not worked out yet.
I've also been doing some digging on SZ viewer and have seen that with the Windows version it is possible to view the raw (hexadacimal) data streaming from the ECU, I may be able to make some sense of that but not managed to get around to setting up on my laptop and getting it into the garage (still a bit cold here), possibly have a go this weekend.

I will invest in a new ELM soon but it just seems a bit hit and miss whether you get a good one. I also bought a female OBDII socket on a pigtail and I'm looking for the Sumitomo socket for the diagnostic connection so I can make up a proper cable, make it a bit easier to plug and play.

 

[GDCobra]

I would be curious to have a screenshot of your diagnosis

It would be interesting to see some more examples of input values, be interesting (to me) to see if I'm the only one seing erroneous readings.

 

[GDCobra]

New ELM on order, hopefully here mid week.

 

[GDCobra]

That looks interesting. Unfortunately I'm going to be out of the loop on this for a good few weeks now. Went out for a cycle ride last Friday, came of the bike and smashed my pelvis I I'm going to be in hospital for at least the next 5 weeks and probably not weight bearing even after that. I'll look forward to updates on this project thuogh

 

[GDCobra]

You do realise, MOTORcycles are better for your health.....

Damn, thats not good..... broken or cracked ? Hope you heal up quicker than they say

It’s pretty well mullered the socket in the pelvis with multiple fractures. It is crazy when people point the finger at motorcycles as dangerous and then I have a result like the while it for a pedal!