M1tch
-
Posts
171 -
Joined
-
Last visited
-
Days Won
1
Content Type
Profiles
Forums
Events
Gallery
Blogs
Posts posted by M1tch
-
-
I have now gone back into the initial map before adjusting by 10% and now simply pulled back timing on any of the highly advanced cells - pulling cells with say 35 degrees advance to 25, those with 30 degrees to 20 etc which will be less aggressive but should combat knock so I can get the fueling right, I can then adjust the timing more advanced once thats setup.
I also need to go through to the knock setup again and adjust the gains per cylinder - I just have the 1 sensor and it might be close to some cylinder than others.
I am also going to grab the AFR table from the Monsoon sample map as a base as well - tempted to almost start again with the fueling and timing by dropping in the maps from the monsoon sample map and a 350z timing map and seeing how that performs.
Just looking into the speed sensor, it seems that looking at the logs the DI1 hz is exactly the same as the speed eg 60mph = 60hz, using the help file with regards to speedo calibration the ABS ring has 48 teeth, rear wheel diameter is 600mm eg 0.6m, this means I need to input 2546 as the calibration figure - max is 500. I have just set it to 1 and will see if that works correctly as 1hz = 1 mph.
-
Just been out again, managed to get a partial throttle pull to higher RPMs to get a few cells worth of fueling on the mixture map, seems to still be triggering knock control - do I have an issue with my setup meaning I am getting false positives?
I am unable to upload the log file as its apparently too big at 2.3mb and I can't seem to be able to reduce the file size enough to be able to upload it on here
I have taken a screenshot of part of the log, it shows that knock control is enabled with the knock value being well under 200 vs the 600 threshold, I have also attached the current map with a few fueling adjustments and 10% timing taken out of it but its still activating knock control.
I have also taken another screenshot of the knock control active with very minimal timing advance - map sensor is showing full vacuum though even at 70% throttle.
What else should I try? Shall I pull additional timing from the map and try it again - perhaps a max of 25 degrees advanced? Although knock control is active at just 10 degrees ignition.
Maybe I am still getting false positives with the knock control setup - I have set it at 7KHz narrowband - should I perhaps set it to 8khz?
I have also just noticed that there is an error value set on the MAP sensor of 80KPa - I am higher than this so perhaps its detecting an error? MAP sensor is correctly setup to be between 0.5v and 4.5v and between 0Kpa and 516Kpa (75 psi).
280718 10% less timing initial adjust.pclr
-
I have just got back from some logging, I can't get any logging on the mixture map above 4k rpm it seems, just checking the initial log it seems that the knock threshold has been reached and I am guessing its not going to log anything in the mixture map if the knock control is working?
Looking at the initial log it seems that at 4k rpm I am at 99Kpa on the MAP sensor with 11 degrees ignition advance which really doesn't seem that much for it to knock considering it increases to 25 degrees advance if it wasn't at WOT and at say 80 KPa.
Checking the logs and the knock status it seems to show that the knock control is active at 1.8k rpm and 52KPa at 10% throttle - looks like it was at around 29 degrees advanced which might be why but the knock level at that point is only 80 (with a threshold of 600).
The knock control is also active until 5k rpm but then locks out due to high RPM - should I look to reduce the RPM lockout on the knock control or is it a case that it doesn't really work at the higher RPMs?
I think the next steps for the map is to pull timing out of the map and then check the knock readings on the next logged run - will then be able to get the fueling sorted on the mixture map (if it is indeed the knock control that is stopping things getting logged) and I can then advance the timing again once the fueling is happy.
The MAP sensor seems to be correctly configured so as far as I know the data is correct, wheel speed isn't correct and I am unsure how to fix that.
I can't seem to be able to upload the logs as they are around 4mb each - will look to reduce timing by 10% across the whole map and use that as a starting point and see what that does, once I get the fueling sorted I can then start sorting ignition etc.
-
Will be taking it out for a drive in a bit to get some logging data - one thing I am unsure about is the wheel speed and how its setup, the speed signal coming in from DI1 which has been added (and put in as LH front) but as far as I know it just pins at a very high speed, it currently has filtering off but I am unsure if it needs to have a pullup resistor turned on, active edge or the calibration. I am actually unsure as to where the ECU is getting the speed signal from but I am guessing one of the ABS wheel sensors - is there any way I can work out how to get the speed setup logged?
-
1 hour ago, JMP said:
you need to check how you've got your ECU logging configured along with the mixture map configuration. Also keep in mind you have the math function available for single cell tuning by pressing the M key when a cell in the fuel table is selected
Will check the settings, I am actually using a tablet at the moment but might look to get a bluetooth keyboard for it as I think it might be a tad easier to have a physical key to press when bringing up shortcuts etc.
-
9 hours ago, JMP said:
I'd avoid using the auto tune function if you're not using a dyno to progressively work through the map. On the street it will just make a mess of your map and it looks like you're already getting some very large steps in the fuel map around idle. Mixture map with decent filtering is a better solution for street tuning
Thank you, yeah I noticed that the fueling wasn't that smooth, I just wanted to try the auto tune the idle cells a bit to make sure everything was working correctly - will look to use the mixture map for road tuning as it takes the average vs the number of samples which will be a better method.
Will drive around a bit this weekend now that I think the knock sensor is correctly setup, unfortunately its basically impossible to get to with the engine in but will be upgrading to a Bosch unit on the new engine.
Quick question about the mixture map and logging etc, will the mixture map get updated and logged without the laptop plugged in and is there a limit to the logging in terms of maximum number of samples? I only ask as I drove the car around a fair bit a couple of days ago on the previous setup including some full WOT pulls but didn't see any logging on the mixture map above 4k RPM. Also need to look at the rev limiter as I think its set to a hard stop at 7k rpm which is like hitting a brick wall!
-
I have updated the file to add in some additional rows which have been interpolated etc when I added them, I have also switched off the AC request in DI3, I have also rechecked the MAP sensor settings and pressed the MAP calibration option as well.
I have also changed the knock sensor to 7Khz as this is closer to the correct knock value and it seems to be reading a slightly more sensible figure.
I have started adjusting the map using the auto tune for the idle cells, will look to log additional data soon so I can adjust the fueling via the mixture map but it seems to all be coming together - just need to work on the wheel speed sensor next I think.
-
3 hours ago, Adamw said:
The fan is on because the AC request is active - at least in the log a couple of posts up.
Hmm thanks for that - will check to see how its setup in the configuration then - the car doesn't have aircon so might need to chase down how its getting that signal. Looks like its been pinned to DI3, will look to simply switch that off in the ECU and it should hopefully then stop the car trying to idle so high.
-
4 hours ago, Adamw said:
Something is still wrong with your MAP sensor calibration if you are seeing -99Kpa under normal driving, that would be very unlikely with a car engine. You may need a -80 or -85KPa row however.
I will go in again and see what might be the issue, I did change the sensor to PSI in the setup rather than setting it to 517 KPa (as the sensor is in PSI for the calibration) - perhaps I am having an issue with it storing to the ECU, will plug my tablet in directly to the ECU and change it rather than tweaking a save file and using that. I will add in a few extra rows as well, will copy the row below up and perhaps adjust it by say 2% to run richer than needed, the mixture map can then pull out fueling if needed.
9 hours ago, JMP said:looks like you'll need to do some configuration of the knock control setup, you've got cylinders reading way over your knock target at idle.
To add another row/column to any of the tuning tables, select the table and press X, then hit the green plus icon for the axis you want to add to and type in the value you want.
I will check to see how its setup - its currently on the OEM narrowband sensor and I think its set to 12khz in the Link, plan to swap to a Bosch doughnut knock sensor at some point though - guessing it might be another configuration issue on my side, I am just using the settings from a 1zz basemap - will look into which options I have for the knock sensor. I am tempted to check the maximum PFC ignition advance I had without any knock and use that as a maximum, think this map is probably a few degrees more advanced. I am also guessing that the knock control is incorrectly pulling timing everwhere which would give me the feeling of the car being slightly sluggish. I believe the OEM part number is 89615-12120 - trying to find out what I need to configure it as. Just using a knock calculator with regards to knock vs piston diameter, it is suggesting that the knock frequency will be at 7.3khz - so perhaps the knock sensor is set too high meaning that its picking up all sorts of other vibrations.
Still having the issue where the cooling fan is running at only around 80C whereas I believe its set to come on at 95C - also had the car idling at 1.5k rpm when fully warm as if its sensing there is extra load such as aircon or something running on the car.
Getting closer to getting the Link setup and working properly - I can also check the correct values via the OBD port, it would seem that the 3Ohm setting for the pullup resistor is about right - will run both side by side next time and compare (using the torque app and OBD2).
-
I have also just noticed on the fueling table that the MGP only goes to -70 KPa whereas the logging is showing it dropping to -99 KPa (full vacuum) - am I able to add in additional rows into the fuel table? The basemap was originally for a turbo 1zz so I guess they didn't need that part of the map as they would have been on boost at that point?
Also seem to have a fair bit of logged knock which I am guessing the knock control would be pulling timing out of the cell etc comparing the ignition advance between this map and my PFC the timing is slightly more advanced than I have tuned the PFC to - max advance on this map is 35 degrees, whereas I had around 32-32 degrees of advance on the PFC with low knock. Will get the fueling sorted out first though - so far using the mixture map and adjusting the fuel table I am pretty happy with idle and cruise as well as light throttle, will keep driving it around a bit under different loads and keep checking the mixture map to see which cells need adjusting.
Is there a way to have the check engine light flash if the knock protection is being activated at all?
-
On 7/25/2018 at 5:39 AM, JMP said:
I'd get it running and let it warm up, then adjust the master so the fueling is roughly right, then you can start the fuel map tuning. The 1zz turbo I'm currently working on has Deatschwerks 60lb injectors on the stock returnless fuel system and it needed a master of around 8ms for the idle cell values to be around 30 for a petrol AFR of 14.4 in traditional fuel mode
Perfect, thanks for your help on this, great to know what I should be aiming for in terms of the right ball park - I have found that on the stock injectors it seems to be ok at 19ms which is the same on the basemap I started with - it does say to start with the master fuel first then adjust the map.
I have just been out for a drive with it all connected up, had the closed loop off but I did have the autotune on for a bit to adjust some of the idle cells etc. It also seems that the speed sensor is incorrectly setup as it seems to pin at around 4000mph
I think the temp sensors look about right, was at around 85c on the driveway after the drive - although the fan was on, its odd as its set to come on at 95C - engine was fully up to temp during the run on the temp gauge so might have been the stock ECU overriding something!
Also need to knock the idle down slightly - looks to be set at 1k rpm, will drop it down to say 900 or 850 I think which is more normal (and I think part of the MOT requirements).
-
1 hour ago, JMP said:
you didn't have the either CAN channel setup to receive lambda from the AEM. I followed the instructions in the Link help file for the AEM gauge and set it up on CAN2 to receive User Stream 2, as your config already has some stuff setup to transmit on CAN1. Don't use closed loop fueling while you're trying to dial in the fuel map, you'll just end up chasing your tail. Only enable closed loop once the fuel map is fully tuned.
Thank you so much, will give it a go tomorrow - just happy I managed to get the new pins crimped and installed without spending a fortune on a crimper! Slowly chipping away at the setup, learning quite a lot as I go. I did check the help file for wideband in general rather than the AEM setup specifically, have turned off the closed loop, shall I knock back the master fuel as well as now it's reading the correct part of the fuelling map with the correctly setup MAP sensor?
-
I have now made the adjustments to the map to correct the error with me setting up the 5 bar MAP sensor, I have also repinned the connector to move AN Volt 3 and 4 to Temp 1 and Temp 2 for the coolant temp and air temp - using the Bosch curve and with the 3k3 option which seems to have done the trick.
I have also wired in my AEM X Gauge into the CAN H and CAN L ports on the connector - however I can't work out how to enable the ECU to show the AFRs, I have set the closed loop to Stoich wideband but can't see the wideband signal coming though for me to be able to do any autotuning - also the AFRs are slightly too lean at the moment so I guess I need to adjust the master fuel? The CAN signal from the AEM wideband is pinned to DI10 and DI9, however I can't seem to see an option for a wideband in the drop down? I am guessing I need to go into the CAN setup and select the CAN2 option - guessing the other port on the actual ECU is CAN1?
-
6 hours ago, Adamw said:
It is not so much the temp sensor calibration that is unknown, it is the value of the internal pull-up resistor in the OEM ecu that is unknown. You cant change sensors as this will upset at least the dash gauge etc. Those graphs above are closer to the Bosch Std NTC curve than the GM/Delphi curve you are suggesting.
Since you have a Fury with configurable temp pull-ups you can move the wires to temp inputs if you like and change the software pullup value (external) until they read close. I suspect "3K3 external" will probably be the correct one.
Ah yes, that makes sense, need to keep the same sensor as its correctly showing on the dash - my Xtreme is setup as a piggyback as some of the functions of the stock ECU can't be run via the Link ECU, I will move the pin over to one of the AN Temp pins whilst I am wiring in the wideband O2 and then configure it to have the Bosch Std NTC curve but with a different pullup value.
The good news is that I can tell that things like the cooling fans are working and coming on when the ECU notices a higher temperature as they clicked on when the engine was warming up (although reading hot on the Link) - will see if there is a way I can read off the temp on the gauge vs the Link - MIGHT be able to use an OBD2 reader to read off the stock ECU whilst the Link is running the other functions as the dash gauge is correct.
-
46 minutes ago, Adamw said:
The unknown with the temp sensors is the value of the pull-up in the stock ecu. Judging from the example temperatures you gave above Im going to hazzard a guess that the pull-up is about 3Kohm. Can you try changing your cal 1 set up to Volts rather than ohms and input the numbers below.
Thank you - will give it a try, I have a spare MAF and temp sensor in the garage so will try and look up the part numbers - might just be simpler to go with a known sensor and values though but will try and get things sorted using the stock sensors - thank you so much for your help so far on this!
Edit - just looking at alternative coolant sensors online, there is one listed with a resistance of 2080, 280 which i am guessing is the max and min resistance of the sensor - tempted to simply swap out the temp sensor though as it should be fairly easy to get to on the engine although it does seem to have a different connector typically!
I think I have found something that might be of use - not sure if it matches off with any other sensor though:
Water temp sensor
Intake air temp
I *think* these are the calibrations I am after - seem to be the same as GM:
Degrees F - Degrees C - Ohms
-40º / -40 / 100,700
0º / -18 / 25,000
20º / -7 / 13,500
40º / 4 / 7,500
70º / 21 / 3,400
100º / 38 / 1,800
160º / 71 / 450
210º / 99 / 185Looks like the air intake temp sensor might also be matching the GM data:
http://pe-ltd.com/assets/air_temp.pdf
Sounds like it might be easier to repin the ECT and IAT sensors to the AN Temp pins rather than the AN Volt pins that they currently are so I can use the 'off the shelf' calibrations rather than a cal table.
-
I have changed the MAP cal table so its now at 0.5, with regards to the temp sensors, I might look to swap the pins over to the AN temp inputs, although I can see that its using a cal table and AN volt at the moment - seems quite odd that the PnP harness would use those inputs.
The sensors are Denso so will try and find the calibration figures for them, I do have a intake air temp sensors still to install (instead of using the MAF) which has the calibration data for it so might see if I can get that installed and use that instead - might be able to use the wiring on the stock plug to wire it in as a simpler solution to get me going.
http://www.efi-parts.co.uk/index.php?productID=181
I might also look to get a coolant temp sensor as well to replace the stock one if that makes things easier as well:
-
8 hours ago, Adamw said:
Map cal table is wrong. I’m replying on a phone so can’t do a screenshot but input A should be 0.5v, not 0v
Thank you, will sort that out, just need to work on the calibration of the temp sensors. Still have the MAF attached which is being used for the air temp, I do have another air temp sensor to use at some point. Wanted to start off with how it's been wired together by Boomslang etc.
-
I have now installed a new AEM MAP sensor - its a 5 BAR one as the 3.5 BAR was out of stock so I have adjusted the sensor calculation for the different scale vs voltage.
Good news is that the car fired up! I had it idling with the master fuel at around 6ms, it was initially set to something like 19 which basically had loads of fuel and smoke out the back as it was running richer than 8:1.
I have my wideband plugged in, however not yet inputting to the ECU - the open barrel crimp tool is on its way but wanted to get it fired up initially.
I have attached a log for my initial start up, I can see the sensor calibration still isn't quite right as its showing a high ECT when the engine is cold as well as a high intake air temp which I think is messing with things - its showing a ECT of 67c with a cold engine and 61C intake temp rather than mid 20s.
At 6ms on the master fuel the AFR on cold start was at around 11.5:1, seems to go lean and stall out if I blip the throttle so I am guessing that the cold start enrichment is keeping it running - once I have the wideband O2 plugged into the ECU I should be able to use the autotune option. The basemap is based off a tuned boosted Toyota Yaris (so would only be using the off boost part of the map!).
Thank you so much for everyone who has helped so far, getting there but I just need a few bit of help to get the car idling and running as it should do - not taken it out as of yet as I know the sensors aren't calibrated correctly as well as not having the AFRs logging at the moment.
-
16 minutes ago, Brad Burnett said:
The first one you listed there will be the best bet out of those 3
https://www.amazon.co.uk/Qiilu-Professional-Terminal-Harness-Crimping/dp/B076CJL1FW/
Thank you, will get them ordered, will check the lead time on it first though.
-
I have asked around at work to see if anyone has a pair, I am over in the UK so might be a bit more limited in terms of some of the options, I have found the below - are any of them any good (sorry they are all Amazon it seem, not a site I ever want to use)?
1.25mm > 6.3mm
https://www.amazon.co.uk/Qiilu-Professional-Terminal-Harness-Crimping/dp/B076CJL1FW/
Similar but better quality:
https://www.amazon.co.uk/Barrel-Terminal-Crimper-DELPHI-Terminals/dp/B01A6U5242/
This one has a set of dies so might be handy:
https://www.amazon.co.uk/Voilamart-Ratcheting-Professional-Non-Insulated-Screwdriver/dp/B07F9VBMW4
-
56 minutes ago, Adamw said:
This pic is not the right tool but it was the clearest pic I could find online to illustrate (Similar to the T10 but the round cavity is too big on this tool). Ideally, you will have both a round cavity about 1.8-2.0mm and a couple of B cavity size options.
And here is the Hozan P706:
Thank you, will look to get a set asap
-
7 hours ago, Adamw said:
That one is not really suitable - that would be a closed barrel crimper. Even though CAN is probably not a highly critical connection, on these particular connectors the crimp needs to be done fairly nicely otherwise the pin doesnt actually fit into the housing after crimping. About the lowest cost crimper I have seen that will do a reasonable job of these superseal 1.0 pins is the T10 micro-crimper. You need an "F" or "B" crimp type cavity for the conductor crimp and a round cavity for the insulation crimp. https://www.racetronix.biz/product.asp?ic=199-t-10
Hozan P706 is another low-cost tool that can do them well.
Thank you for the info, will go and buy a crimper and do it properly, will probably need to crimp a few connectors in the future.
What do I need to look for when ordering? Just open barrel crimper? Are there different types?
-
Are these ok to use?
-
8 minutes ago, Brad Burnett said:
yes you have to push the white tab
crimp the terminals with an open barrel crimper standard is not ok.
open barrel crimper should be available at like a frys or computer/electronics store.
Thank you, will go and find an open barrel crimper, I am still awaiting a barb adapter to arrive for me to be able to wire in the Link ECU with the MAP sensor that has just arrived so have a bit of time!
Will check to see what my wire stripper has and if its ok to use - might actually be the sort I need.
Installing monsoon into 13b pp help
in G4+
Posted
Do you know if there is anywhere I can view the help file offline? I sometimes want to check something during lunch at work but can't as I don't have PClink on my PC etc, I do however agree that its very helpful.