Knock count firmware
Moderator: Matt
Knock count firmware
This thread will contain the ENT files for knock count firmware.
Several HCR32 Type 2 boards posted out have the Rev15 firmware but used an older patcher program which doesnt do the counting. If you open the knock display window and tap the knock sensor and it doesnt count, then you have one of those boards. Please email me and I'll organise an exchange over provided you can get the other board back to me ASAP since my stocks are limited
I've got another report of Type 1 board (S13 CA18) counting by itself and causing running issues which I'm to investigate tonight
HCR32 updates are attached
Several HCR32 Type 2 boards posted out have the Rev15 firmware but used an older patcher program which doesnt do the counting. If you open the knock display window and tap the knock sensor and it doesnt count, then you have one of those boards. Please email me and I'll organise an exchange over provided you can get the other board back to me ASAP since my stocks are limited
I've got another report of Type 1 board (S13 CA18) counting by itself and causing running issues which I'm to investigate tonight
HCR32 updates are attached
- Attachments
-
- hcr32_rb20_knock_count_rev15_fix.zip
- HCR32 Rev15 knock count fixed
- (182.48 KiB) Downloaded 445 times
I've got another CA18 ECU running tonight (fresh ECU) with the image from the S13 CA18 ENT directory and seems to be running fine
Detects knock as seen in the picture. I'll need to work out what is going on with your ECU Ashley since code is functioning correctly
Detects knock as seen in the picture. I'll need to work out what is going on with your ECU Ashley since code is functioning correctly
- Attachments
-
- s13_ca18_44F02_rev15_knock_feedback.JPG
- S13 CA18 44F02 knock trace
- (260.41 KiB) Downloaded 10824 times
-
- s13_ca18_knock_ent.zip
- S13 CA18 ENT files (generated 3 April) used 44F02 from this pack for testing
- (211.08 KiB) Downloaded 444 times
Hi Matt
I have attached two log file below, I start logging once I start the motor and you will see the Knock registers on the logfile and only go away after I clear them, I have used your Knock monitoring tool, Cell Flagging and alarms and nothing works I the only knock I might get is when the car goes into a server miss when climbing onto boost.
I don't know it this is related to my TTP/min. TTP/MAx issue but the AFR are at 11 when the car starts miss. I Have tried switch the TTP around with no results.
I have attached two log file below, I start logging once I start the motor and you will see the Knock registers on the logfile and only go away after I clear them, I have used your Knock monitoring tool, Cell Flagging and alarms and nothing works I the only knock I might get is when the car goes into a server miss when climbing onto boost.
I don't know it this is related to my TTP/min. TTP/MAx issue but the AFR are at 11 when the car starts miss. I Have tried switch the TTP around with no results.
- Attachments
-
- NIStune_2009-04-16_1627_02.csv
- Log File
- (1.2 MiB) Downloaded 455 times
-
- NIStune_2009-04-19_1532_42.csv
- log 2
- (429.96 KiB) Downloaded 442 times
knock count
Matt.
Could you give me some idea as to what the knock count actually is.
I have done some searching as obviously not all OEM management systems read the same way as far as frequencies of knock and limits of voltage outputs go. One method i found was using actual knock counts, but the counts are relative to voltage outputs of the piezio crystal ie 0 knock was 0 volts and 1020 knock count was 5 volts. What should be our limits and what limit is required to turn cells red?
Another thing while on this, i had primary fuel and primary ignition maps open, both had knock detect checked, but the fuel map showed more red cells than the timing map did.
Everything seems top be working great.
Thanks.
Could you give me some idea as to what the knock count actually is.
I have done some searching as obviously not all OEM management systems read the same way as far as frequencies of knock and limits of voltage outputs go. One method i found was using actual knock counts, but the counts are relative to voltage outputs of the piezio crystal ie 0 knock was 0 volts and 1020 knock count was 5 volts. What should be our limits and what limit is required to turn cells red?
Another thing while on this, i had primary fuel and primary ignition maps open, both had knock detect checked, but the fuel map showed more red cells than the timing map did.
Everything seems top be working great.
Thanks.
HCR32 has a knock sensor board
The knock sensor passes the noises through into this board via pin 2 making a variance on 2.4 volt input
This board then does frequency filtering and outputs the signal as a square wave on pin 6 of the knock board
The ECU takes this input in and then adds it to a raw knock counter. This is where we currently pick up knock has occured and report it via consult
The ECU from here does some further checks against RPM, TP, battery voltage and things like if the engine is cranking before using it internally with a separate 'conditional' knock count. This count is used to pull timing and depending on how high it gets, then triggers the ECU to use knock maps
Your Nistune board uses the knock filtered output of the ECU and then counts the amount of knocks per fuel/timing map cell inside the software. I'm still to determine how many knocks to filter before reporting. I prefer to use the AFR trace screen for now to count number of knocks
I'll have a look at the fuel / timing maps to see the 'red' differences here when I get a moment. Still chasing up Z32 last night and we have got the post conditional knock check which may be more suitable for reporting but at long at it doesnt get software filtered out too much
The knock sensor passes the noises through into this board via pin 2 making a variance on 2.4 volt input
This board then does frequency filtering and outputs the signal as a square wave on pin 6 of the knock board
The ECU takes this input in and then adds it to a raw knock counter. This is where we currently pick up knock has occured and report it via consult
The ECU from here does some further checks against RPM, TP, battery voltage and things like if the engine is cranking before using it internally with a separate 'conditional' knock count. This count is used to pull timing and depending on how high it gets, then triggers the ECU to use knock maps
Your Nistune board uses the knock filtered output of the ECU and then counts the amount of knocks per fuel/timing map cell inside the software. I'm still to determine how many knocks to filter before reporting. I prefer to use the AFR trace screen for now to count number of knocks
I'll have a look at the fuel / timing maps to see the 'red' differences here when I get a moment. Still chasing up Z32 last night and we have got the post conditional knock check which may be more suitable for reporting but at long at it doesnt get software filtered out too much
Morning Matt...actually before i do that... please check a few things. i was doing some more work last night
If you disconnect your knock sensor do you still have the problem?
What is the resistance in ohms of your knock sensor?
I can’t check the ohm of my detonation sensor because of my custom intake manifold and clearance issues but Unplugging it results in a DTC code on NIStune and the car feels slugish