Pete_89t2
-
Posts
178 -
Joined
-
Last visited
-
Days Won
4
Content Type
Profiles
Forums
Events
Gallery
Blogs
Posts posted by Pete_89t2
-
-
First some background - I've been running modeled fuel equation mode with 2D injector dead time tables (voltage vs. dead times specified at my base diff. fuel pressure = 43.5psi) for some time now on my Link G4+ managed FD RX7. The primary & secondary injectors are different, so obviously the 2D dead time tables for the primary & secondary injectors were different too. Since I'm running a fuel pressure sensor, and I recently obtained what I hope is good dead time data for my injectors, I wanted to experiment with 3D dead time tables for both primary & secondary injectors to get a better tune that can adjust for minor changes in differential fuel pressure.
I went ahead and switched to 3D dead time tables in the main injector setup menu, and populated the now 3D Injector Dead Time Table with the correct data for my primary injectors. However, it appears I lost the "Secondary Injector Dead Time" table control that used to exist under the "Staged Injection" menu, so I can't enter the 3D dead time data for my secondary injectors. Previously, I had a 2D Secondary Injector Dead time table there, which is now gone.
What do I need to do to fix this? BTW, I'm making these changes offline with the PCLink software (e.g., make my changes, save tune file & download to Link ECU later), not live & connected to the ECU - is that perhaps the root of my problem?
-
Just as an FYI, not sure if the the Link support folks are aware of this, but I ran into a minor bug when updating my firmware from the 5.6.8.3668 version I had to the current 5.6.8.3671 version that comes with the current version of PC Link (English/Spanish) that I downloaded earlier this evening.
Basically after you connect PCLink to the ECU, and start the firmware update, the update process tells the user that you already have the current/latest version of the firmware (.3671), even though that is not true. Anyway, since I knew I had the new version (.3671) with my freshly downloaded & installed version of PCLink, I selected the "Reinstall firmware" option, and as expected the firmware version reported by the ECU after the process completed changed from .3668 to .3671
Looking forward to doing some test driving tomorrow to see how well CL Lambda control works after the update!
-
-
I'll preface this by saying I'm still a newbie at tuning, and this is the 1st time attempting to try closed loop lambda control on my car. Car in question is a modified single turbo FD RX7 with a DBW throttle. My fuel map can still use some more fine tuning, especially in the boost regions above ~7psi (WG springs), but the fuel & conservative spark timing it has now allow it to run well from idle to about 6500RPM within that MAP envelope.
Today while testing the CL Lambda controls, I noted a few hiccups driving. The first is the idle isn't as smooth & stable as it was before, and the 2nd more pronounced effect was what felt like a fuel cut under relatively light acceleration. Here's links to the car's current tune and log file that captured the hiccups encountered.
https://drive.google.com/file/d/1IYrXmKKCXjMkhQkJUH2geDKVL2SyBhSY/view?usp=sharing
https://drive.google.com/file/d/1hddxt80bBJFXfNsUnCzWR3EaLCWcf3z_/view?usp=sharing
Upon looking at the log, I'm a bit stumped as to what to make of it? Shortly after time index 2:30, you can see CL Lambda correction & status data go from a small +/- correction % and the normal "Incr/Decr. fuel status" to over 6000% correction and "Min Clamp" status. This was coincident with driving off out onto the main road, but I was accelerating rather lightly, as TPS Delta was reading below 2.2% during the log transient, and TP(Main) was at about 6%.
If anyone can take a look at the log & tune and help me understand what's going on I'd appreciate it, thanks!
-
Thanks Adam, that's a great explanation. I was aware of how a Wankel requires 1080 degrees of crank revs to complete a full 4-stroke Otto cycle, hence the question. Glad it uses the same strategy the Mazda ECU uses.
-
Greetings, I'm curious to learn a bit more on what goes on behind the scenes when the "Rotary Only" limit mode is engaged. Per the help file (see excerpt below) it looks like it combines a Ignition % and Fuel % cut in its implementation, but doesn't say much else with regard to how the cuts are implemented.
Limit Mode
The limit mode is used to turn the system on and off. When turning the system on, there are four available limit methods.
- OFF - Limiting is turned off.
- Ign %Cut - An ignition only progressive cut.
- Fuel %Cut - A fuel only progressive cut.
- Rotary ONLY - (Ign+Fuel) A special ECU controlled fuel and ignition limit for rotary engines only. MUST have Injection Mode set to Sequential Staged.
Assuming that I'm using the default Rotary Only limit mode (i.e., advanced mode OFF, and dual MAP limit OFF), can someone explain what the ECU does in sequence when the limit is engaged? For example, is a progressive ignition cut applied first followed by the fuel cut? Or are the ignition/fuel cuts simultaneous, or are there any other variants involved, like perhaps alternating the ignition/fuel cuts between the 2 rotors?
-
Thanks, looks like I'll settle for just for the EGTs then, and look into a good CAN EGT amp that can do at least 2 EGT TC's.
-
2 hours ago, Confused said:
This ECUMaster item is relatively small and cheap.
https://www.ecumaster.com/products/can-switch-board/
However, the G4+ CANBUS is a bit limited for analog inputs - but if you just want to log them (and don't mind that the channel names might not be totally accurate), then this is a relatively cheap way to do it.
However, don't think it'll do EGT directly...
Thanks, that should do most of what I'd need - interested in only logging, so I can live with funky channel names. I realize I'd need a signal conditioner box to convert the typical EGT thermocouple outputs into a proportional 0-5VDC analog signal for the G4+ or ECUMaster to make use of.
-
Haven't really dug too deep into this, but I was wondering what extra hardware I would need to add perhaps another 4 analog inputs to my G4+ Fury for logging purposes? I'm assuming I'd need a box that can accept some # of additional analog channels and put that data on the CAN bus?
Background is that my current install (FD RX7) already makes use of all the analog inputs available on the Fury, and I've been toying with the idea of adding 2x EGT sensors and perhaps EMAP sensors, and would like to capture that data in my logs.
-
7 minutes ago, Adamw said:
The default control range if you have advanced mode turned off is 10kpa.
Thanks Adam, that explains it! I'm using the "Rotary ONLY" setting for the MAP Limit Mode, and Advanced Mode and Dual MAP Limit are both set to "OFF", so I'd assume the default control range is still 10Kpa, correct?
-
As an update to this topic, I've double checked the wiring harness wiring to pin 4, and I have good continuity and correct voltage on the pin when energized, so I doubt that power supply is the problem. I am now seeing Lambda 1 Error change from "16 Heated Too Long" to error "6", after about 30 minutes of driving instead of error "24". Lambda status still shows "Operating", and the lambda 1/Lambda average data looks OK.
Similar to error #24, the help file states error #6 = "Internal control error, no user actions can be taken to correct. Report this to your Link distributor." What does error #6 indicate?
At this point I'm assuming my WBO2 sensor is bad, so I went ahead and ordered a new one, hopefully that will fix the problem.
-
Some background - As I don't have access to a dyno, I've been doing all of my tuning to date on the road. To do so safely with less risk to my engine, I've been setting MAP limits slightly above the boost level I'm trying to capture log data to tune up to.
Now that I'm getting my tune up to the neighborhood of my waste gate's spring pressure (7psi), I'm finding that the MAP limit cuts that I programmed cut in at approximately 1 psi of MAP lower than it was set. Specifically, I had my MAP limit set to 23.5psi, and the MAP cut went active at only 22.5psi MAP in the logs.
Just curious, is this behavior by design, to err on the side of engine protection? If so, it's easy enough to raise the MAP limit a bit.
-
1 hour ago, Howard Coleman said:
am i correct that the split timing numbers reported in a log are not the actual timing but just reflect the split settings?
for example:
i am at 8.2 lead and my split shows minus 11 on the log. my actual split timing event at this point would be 8.2 minus 11 or minus 2.8.
please confirm. thank you
That is correct for the G4+ series of ECUs, probably also true for G4X, but I'd wait for someone from Link to confirm.
-
Perfect, thanks again!
-
30 minutes ago, Adamw said:
Their instructions are wrong. Extended stream 1 should be on ID 50, and extended stream 2 should be on ID 51.
Thanks Adam. It looks like I had ID 50 for extended stream 1, but per GaugeArt's instuctions, I only had 2 channels setup (standard on ID 56, extended on ID 50 for both streams). How/where would I change extended stream 2 to ID 51 - would I just add a 3rd channel (currently OFF) under the "Mode" tab, and assign it to ID 51?
-
I purchased one of these GaugeArt CAN bus gauges for my FD RX7, which is running a Link G4+ Fury ECU.
Problem I'm having is the gauge refuses to display some of the extended channel data GaugeArt claims it should, for example Lambda target. Tech support folks at GaugeArt has been telling me for months now that they have been working with Link to come up with an updated set of configuration instructions that should fix my problem, but they have not delivered yet. Was hoping maybe someone here has figured out a solution for this. I simply want to be able to set up a 2 parameter display screen that shows Lambda 1 (a standard channel) and Lambda Target (an extended channel) on the same page. When I try to do that according to the GaugeArt instructions, Lambda 1 displays properly, but I get a constant 0.000 for Lambda Target.
Anyway, here's a link to the standard and extended channel list this gauge is supposed to support and display: https://gaugeart.com/instructions/gaugeart-can-gauge-supported-ecu-data-channels/supported-ecu-channels-vipec-link-g4/
And here is a link to the instructions GaugeArt provides its users for them to configure a Link G4+ to work with their CAN bus gauge: https://gaugeart.com/instructions/vi-pec-link-g4-ecu-setup/
I've followed those instructions to the letter, with one exception - Since I wired my CAN gauge to DI9/10 (CAN2 module), I substituted CAN2 module for CAN1 module in their instructions. The GaugeArt gauge seems to work fine that way otherwise, and I can configure it to display everything it should, EXCEPT for the following extended channels:
- Lambda Target
- Lambda Correction 1
- Lambda Correction 2
- Boost Target (kPa / PSIg)
Thanks in advance!
-
1 hour ago, koracing said:
I would personally use Excel to do this and use excel to do the tune refinement as it can be set up to use closed loop corrections as well where mixture map won't.
Not sure I follow; are you saying that you can open & manipulate PCLink log files in MS Excel? Please explain, didn't know that was possible.
-
3 hours ago, Adamw said:
The error 16 is usually related to power supply. How long and what size is the wire to pin 4 on the lambda sensor? If that appears acceptable for say 8A max, I think it would pay to try moving that to a different power source.
From pin 4 it's an 18AWG, M22759/32 spec wire run, maybe 3 feet long total to the WBO2 sensor from where the wire connects to the 14AWG switched ECU/power bus inside the wiring harness in the cabin. Didn't realize the heater circuit pulls as much as 8A, that could be my problem.
-
UPDATE - I reviewed a 2nd log, with a longer duration (about 1 hour) but also starting the car from overnight cold. In that log, the behavior is essentially the same as my post above, but at approximately 38 minutes in, the Lambda 1 Error changed to "24", where it remained for the rest of the log, and Lambda Status remained at "Operating" as well for the rest of the log.
-
5 hours ago, Adamw said:
Error 24 means the current regulator has reached the limit. It would typically indicate a failing sensor.
It lambda 1 "run when stalled" set to no?
Adam,
Just confirmed I have "run when stalled" set to "no".
Looking at the last log I have from a cold overnight start, below is the gist of the behavior. On this log, I only had the error #16, "Heated too long" show up, never had an error #24 show up.
@ 19.4 seconds after start, Lambda 1 status = "Heating", Lambda 1 Temp = 1225*F, Lambda 1 Error = "OK"
@ 19.5 seconds after start, Lambda 1 status = "Diagnostics", Lambda 1 Temp = 1236*F, Lambda 1 Error = "16 Heated Too Long"
@ 21.7 seconds after start, Lambda 1 status = "Calibration", Lambda 1 Temp = 873*F (why the drop?), Lambda 1 Error = "16 Heated Too Long"
@ 32.5 seconds after start, Lambda 1 status = "Heating", Lambda 1 Temp = 1144*F, Lambda 1 Error = "16 Heated Too Long"
@ 41.7 seconds after start, Lambda 1 status = "Diagnostics", Lambda 1 Temp = 1398*F, Lambda 1 Error = "16 Heated Too Long"
@ 43.6 seconds after start, Lambda 1 status = "Calibration", Lambda 1 Temp = 1398*F, Lambda 1 Error = "16 Heated Too Long"
@ 43.99 seconds after start, the Lambda 1 status = "Heating, "Diagnostics" & "Calibration" cycle similar to the above repeats itself a few times until...
@ 1 minute 2.4 seconds after start, Lambda 1 status = "Operating", Lambda 1 Temp = 1416*F, Lambda 1 Error = "16 Heated Too Long"; Lambda status remained at "Operating" for the remainder of the log (about 30 minutes of driving), and lambda temps ranged from 1416*F to 1458*F during that time.
-
I've noticed lately in my logged run times that I'm getting Lambda 1 Error #16 (Heated too long), which is sometimes followed by Lambda 1 Error #24, which the help file essentially says to "contact your Link dealer" as a description. While it's doing this, the Lambda 1 Status is reading as "Operating".
Lambda temperatures when this happens seem normal, and they vary around 1430*F, +/- about 20*F typically. On an overnight cold start, I can see the Lambda temps smoothly & quickly increase from ambient air temp to the 1430*F temps once the heater kicks in.
Is this a sign that my WBO2sensor is failing, and what does the error #24 mean?
-
The subject is the gist of my question. Reason I'm asking is I'm using the Mixture Map feature to refine my tune, and I think the process would be more productive if I could combine logs from multiple road tuning sessions (with no changes to my tune in between each session) into a single large log file. By doing so, I expect I'd be able to hit more load/RPM cells within the fuel map, and have larger sample sizes per cell for the Mixture map feature to work with.
If this feature is not currently available, can it be added to the wish list? Thanks!
-
I didn't get a resistance measurement, but after hooking it up and doing a little testing, it appears this IAT sensor is happy with the "Std Bosch NTC" calibration data that's built into my G4+. With the car overnight cold, the new sensor tracked within 2~3*F warmer than the other IAT sensor I have installed. The new sensor (Triumph) is located in the FD's upper intake manifold, and the other IAT sensor (GM Delphi fast response IAT) is just before the throttle body inlet elbow on the inter cooler piping. After a drive, the two sensors remained in close agreement, trending in the same directions within a few degrees F of one another. As expected, the Triumph sensor (in the UIM) read a bit higher due to its location being more influenced by heat soak.
-
19 minutes ago, Adamw said:
Whats the resistance at room temp ~20deg C?
I'll find that out and report back when the sensor arrives in the mail tomorrow and I plug it in. Might not be 20*C exactly but I'll be able to compare the Triumph IAT (mounted in the upper intake manifold) with the other IAT sensor I have (mounted in the cold side IC piping), and with ECT & Oil Temp readings. When the car is left overnight cold, all of those readings will typically read within 1~3*F of one another, at the current ambient air temperature.
Secondary Injector dead time table no longer accessable after switching to 3D tables?
in G4+
Posted
Thanks Adam. My ECU is running version 5.6.8.3671 firmware, which I think is the latest available. Here's what my table allocation looks like:
So tables 1 thru 11 are already used up, but tables 12 thru 30 are not allocated to anything. I'm guessing if I plug into the ECU with PCLink, the lock symbols would go away and I'd be able to allocate T12 as the Secondary Injector Deadtime table?