logger problems

ByReaL · Post by **ByReaL** » Mon May 07, 2007 7:19 am

i'll post all the problems here:

1) afther a log file is loaded if you select from teh combobox another parametr (for example from TPS you change to aux1) you have to load again in order to make the graph look ok.

1bis) you open the log player, and in comboboxes you have for example TP, TPS, AFR (wideband) , Aux1 you load a log file, but AFR and Aux1 are not properly loaded, you have t oopen the combobox reselct the AFR and Aux1 and load again the log file

3) TP is not proprly loged, i have attached a file with a run made in 4th gear full thrrotle, when i play that file it shows me that the TP ismostly in the first and sometimes reach the 3rd or 4th load columns, but this is wrong because when i did teh run i could see the how the ecu readed from higher load points

4) log is not right, as you can see from the log file attached there si apoint where the speed jumps from 154 t o180km/h in a fraction of a second (this si imposible), so considering that the TP is not logged to well and this speed jump, i belive there is a problem with the logging module

Matt · Post by **Matt** » Mon May 07, 2007 10:53 am

TP has to do with me leaving debug in hex. I have fixed this in the next upcoming version (yes it stuffed all my logs up this weekend)

I'll have a look at the logging stuff... currently working on that

ByReaL · Post by **ByReaL** » Thu May 10, 2007 6:48 pm

1) in the logged csv file i can see knock fuel flag, and sometimes has the value 80 and sometimes has the value 84, but when it is played from nistune only the knock fuel map is highlighted.

2) i noticed in the logged CSV there is a field with knock, i want to ask if this thing is showed somwere during playback, my sugestion (enhancemet request) is to add to log player also the Knock parameter, so you can see on graphs where knocks ocured

Matt · Post by **Matt** » Fri May 11, 2007 1:38 am

please dont worry too much about the logging flags etc from that version.

knock sensing is in the middle of work (ie not fully functional) so please dont raise bug issues against it yet

i am just making releases of nistune in the interim so that other bugs raised can be fixed. that part still needs to be tested

Any flag values are recorded/playback in the CSV file but they do not have a special item for display. They will be visiable on the consult display window and in the knock display window during playback

I noticed about the list box not updating in the logger (and also the newly added Knock display window for additional display item). This still needs to be fixed and is on my list

I'll have to look into the speed stuff

ByReaL · Post by **ByReaL** » Fri May 11, 2007 5:38 am

i still belive that will be a grate help if knocks will be added also on the log player, so the knock can be viewed on graph.

if i want to see where knocks ocured i have to play all the log and have an eye on the "D" window waiting to see if there is any knoc or not, if there will be available on log player i'll be able to see if knocks ocured just by looking at the graph for "knock spikes" without haveing to play the whole file.

anyway as you said this knock things are "under construction" i'll wait for you to finish the work on this and then i'll do any comments if there will be any.

Matt · Post by **Matt** » Fri May 11, 2007 11:27 am

no probs. i understand the reason why you want to see it on the playback now...

some flag registers for example may have
- TPS Switch
- Air Cond
- Start switch

A register (for a value) is associated with a single colour line... but as you can see from the above there are 3 values which are all on/off. similar with knock sensing stuff

Any ideas how three items at the same time from a single register can be displayed on the graph? Currently the graph is just a single line drawn across the window

Trying to display something like flag registers which are mulitple on/off items is not gonig to work too well

LMQT · Post by **LMQT** » Sun May 27, 2007 6:48 pm

Any news for logging the TP ?
Or is there any older version who log it correctly ?

It's really important for me

ByReaL · Post by **ByReaL** » Mon May 28, 2007 3:40 am

on the last version xxxxxxx19 it works ok

Russ84NA · Post by **Russ84NA** » Mon May 28, 2007 6:00 am

Matt,
I was logging some today, The AFR was logging as I could see it on the graph and it was showing up on the A/F graph window. However, when I went to play it back it dissappeared. It took one of my old logs with it, it just vanished.

Matt · Post by **Matt** » Mon May 28, 2007 1:43 pm

looks like a separate issue.

can you remember what operations you did to get playback? it looks like it may have overwritten it?

LMQT · Post by **LMQT** » Mon May 28, 2007 9:48 pm

TP LOG

LMQT wrote:Any news for logging the TP ?
Or is there any older version who log it correctly ?

It's really important for me

ByReaL wrote:on the last version xxxxxxx19 it works ok

Effectively it's works. Thanks

Matt please, could you tell us the formula used to convert TP in the log towards final TP (the one we can see on the columns in the fuel/timing map) ?

AFR LOG
For me the AFR is well played.
As well in live as by replaying the log.
I've got an AEM if it could help...

ByReaL · Post by **ByReaL** » Tue May 29, 2007 4:24 am

maybe another problem (i say maybe because i do not have untill the weekend comes time to investigate it)

start the logging, let it log for some time, remove the usb cable from laptop (simulate consult disconect), log is lost

Bernardd · Post by **Bernardd** » Tue May 29, 2007 8:57 am

tp = value/256

Matt · Post by **Matt** » Tue May 29, 2007 11:23 am

The AEM works?! Thanks .... I hadn't had a test report from this yet

The TP is split over two registers [16 bit] but then you take the highest register [8 bit] only for the scalers (so as bernard mentioned /256) i can display this on the guage perhaps as two numbers

.... I'll try out the USB disconnect problem

LMQT · Post by **LMQT** » Wed May 30, 2007 4:42 am

Matt wrote:The AEM works?! Thanks .... I hadn't had a test report from this yet

The TP is split over two registers [16 bit] but then you take the highest register [8 bit] only for the scalers (so as bernard mentioned /256) i can display this on the guage perhaps as two numbers

.... I'll try out the USB disconnect problem

Yes the AEM works pretty well : connected this weekend and it's perfect

OK for TP.
For the log, it can stay in this form (it could be usefull on day, who know ....) as long as we know how to retrieve the final TP

But effectively it could be display correctly on the gauge, it's better for an immediat translation ?

Thanks, Jef