Virtual Dyno
01-05-2012, 07:59 AM
#25
Junior Member
Join Date: Jan 2012
Location: Kansas City, KS
Posts: 66
Total Cats: 3
If you will send me a few logs with TIME, RPM, TPS, BOOST, and AFR in them Ill be glad to support them. That doesnt mean the log must contain those exact columns just columns with that data in them. I need an UNEDITED file in CSV or TXT format. You can either post it on the forums for me to get or send them to bradbarnhill at hotmail com. If you send via email explain what they are and such as I get a lot of emails a day.
Thanks,
Brad Barnhill
Virtual Dyno
Thanks,
Brad Barnhill
Virtual Dyno
Reply
2
2
01-05-2012, 02:54 PM
#26
This is what I've discovered so far playing around with Virtual Dyno:
1. Virtual Dyno uses the names found in the first row of a comma-delimited text file to get its parameters. The first row limitation is important. For example, a Megasquirt log writes the software version number on the first row and the parameter names on the second row. Therefore, you need to open that Megasquirt log and delete the first row for it to work with Virtual Dyno.
2. Virtual Dyno names can be assigned to match what is in your file with a pulldown menu (under File=>Options=>Columns and Profiles). The parameters that Virtual Dyno needs to work its magic are:
-Time (Required for dyno plot)
-RPM (Required for dyno plot)
-AFR (Optional, needed if you want the AFR trace below the dyno plot)
-Boost (Optional, needed if you want the Boost trace below the dyno plot)
-TPS (This is listed in the menu, but I can't discern how/if it is used)
3. Because of #2 above, I really wouldn't pay much attention to the "list of supported ECUs" on Virtual Dyno's website. If you are handy with a sheet or text editor program and your ECU's output is human-readable text, you'll be able to use Virtual Dyno.
MS1/MSPNP Instructions:
Here's an example of setting up with a log file from a MSPNP9093 (basically a Megasquirt I):
1. Open log file in Excel and delete first row. New first row should be parameter names.
2. Edit the data so that the pull you want to show is the only data present. This isn't strictly necessary, but you will get a lot of unwanted artifacts on your plot if you don't do this.
3. MS1 plots have "Time" and "RPM" columns. These are used as-is by Virtual Dyno.
4. MS1 plots have the wideband readout in volts under "O2" in Excel column F. To convert this to AFR, you need to know what the volts represent. In my case, I'm setup so that 0VDC = 10:1AFR and 5VDC = 20.8:1AFR. So, I created a new column in my data called "AFR" and used the following equation:
"=10+F#/5*(20.8-10)" where # is the Excel row number
I then opened "File=>Option=>Columns and Profiles" in Virtual Dyno and put "AFR" into the "AFR" option. I had expected that by calling the column "AFR," Virtual Dyno would know what to do with it, but it didn't work out that way. ?? Make sure the check mark is on next to "AFR."
5. If you want Boost plotted in units of PSI, then you need another calculated column. MS1 plots have a manifold pressure readout in KPa under "MAP" in Excel column D. MS1 plots also have an atmospheric pressure readout in KPa under "barometer" in Excel column AC. So, create a new column in your data called "Boost" and use the following equation:
"=14.696/101.3*(D#-AC#)" where # is the Excel row number
Open "File=>Option=>Columns and Profiles" in Virtual Dyno and put "Boost" into the "Boost" option. I had expected that by calling the column "Boost," Virtual Dyno would know what to do with it, but it didn't work out that way. ?? Make sure the check mark is on next to "Boost."
6. I actually prefer plotting MAP in KPa instead of boost. This makes it easy for me to correlate to my spark and fuel maps. To make this happen, open "File=>Option=>Columns and Profiles" in Virtual Dyno and put "MAP" into the "Boost" option field. Done. Couldn't be easier. Again, make sure the check mark is on next to "Boost."
7. There is also a TPS field in Virtual Dyno. In MS1, the thottle position value is called "TP" in Excel column E. I put "TP" into the "TPS" option field. Doesn't seem to have any effect. ??
Hints:
This is no different than a real dyno. To make valid comparisons, you need to eliminate variables so that, hopefully, only the effects of the tuning parameter you are working on are seen. To that end:
1. Do your comparison runs at the same time and place if possible. This is easy if you are just changing ECU parameters (make a run, reprogram ECU, make another run). Harder if you are changing mechanical items. If changing mechanical items, then use the same stretch of road and try to do compared runs under nearly identical weather conditions.
2. Be cognizant of aerodynamics. Windows, top and, for NAs, headlights will change your results. To get a "close" horsepower reading using the Mazda-published drag coefficient, you need to have windows closed, top up (or hardtop on), and NA headlights retracted.
3. Do two runs on the same stretch of road going opposite directions. This helps cut down on the effects of wind and slope. Your comparison plot would be the average of the two runs.
Bottom Line:
Does it work? Heck, yes, it works. Last night, I tried it in earnest for the first time. I wanted to compare two spark maps. The first is based upon the MSPNP default spark map as follows:
The second has a bit more timing, especially at high RPM:
Here's the result plotted by Virtual Dyno:
The additional timing at high RPM is clearly seen as increased torque and HP in that region. MUCH better than a butt dyno and, so far as I can tell, the data seems to be as useful as that from a real dyno. Now, if only I could hold a constant RPM . . . . Nonetheless, this looks like a valuable tuning tool to me.
Now, about the numbers . . . not too impressive. Keep in mind that my headlights were up. I'm certain that if they had been down, I would have gotten 270RWHP+ out of my non-intercooled Greddy.
Possible Bugs for Brad:
1. Need to specify AFR custom column name even if column name is "AFR."
2. Need to specify Boost custom column name even if column name is "Boost."
3. What does TPS do?
4. The auto update of car profiles overwrote a custom profile that I had stored. This was mildly annoying. It would be nice if there were a message that informs the user that new car profiles have been written.
5. I would love to be able to plot an average of multiple runs. That way, I could combine going two directions on a stretch of road. Doesn't seem like it would be too hard.
6. Help leaves a bit to be desired (there isn't any). Feel free to use anything I've written here to get that started.
7. Cool app. Glad to have you aboard. I'm not sure whether you are a Miata guy, but you owe it to yourself to take a drive one. If you're ever in the DFW area . . . .
8. You get a prop!!
1. Virtual Dyno uses the names found in the first row of a comma-delimited text file to get its parameters. The first row limitation is important. For example, a Megasquirt log writes the software version number on the first row and the parameter names on the second row. Therefore, you need to open that Megasquirt log and delete the first row for it to work with Virtual Dyno.
2. Virtual Dyno names can be assigned to match what is in your file with a pulldown menu (under File=>Options=>Columns and Profiles). The parameters that Virtual Dyno needs to work its magic are:
-Time (Required for dyno plot)
-RPM (Required for dyno plot)
-AFR (Optional, needed if you want the AFR trace below the dyno plot)
-Boost (Optional, needed if you want the Boost trace below the dyno plot)
-TPS (This is listed in the menu, but I can't discern how/if it is used)
3. Because of #2 above, I really wouldn't pay much attention to the "list of supported ECUs" on Virtual Dyno's website. If you are handy with a sheet or text editor program and your ECU's output is human-readable text, you'll be able to use Virtual Dyno.
MS1/MSPNP Instructions:
Here's an example of setting up with a log file from a MSPNP9093 (basically a Megasquirt I):
1. Open log file in Excel and delete first row. New first row should be parameter names.
2. Edit the data so that the pull you want to show is the only data present. This isn't strictly necessary, but you will get a lot of unwanted artifacts on your plot if you don't do this.
3. MS1 plots have "Time" and "RPM" columns. These are used as-is by Virtual Dyno.
4. MS1 plots have the wideband readout in volts under "O2" in Excel column F. To convert this to AFR, you need to know what the volts represent. In my case, I'm setup so that 0VDC = 10:1AFR and 5VDC = 20.8:1AFR. So, I created a new column in my data called "AFR" and used the following equation:
"=10+F#/5*(20.8-10)" where # is the Excel row number
I then opened "File=>Option=>Columns and Profiles" in Virtual Dyno and put "AFR" into the "AFR" option. I had expected that by calling the column "AFR," Virtual Dyno would know what to do with it, but it didn't work out that way. ?? Make sure the check mark is on next to "AFR."
5. If you want Boost plotted in units of PSI, then you need another calculated column. MS1 plots have a manifold pressure readout in KPa under "MAP" in Excel column D. MS1 plots also have an atmospheric pressure readout in KPa under "barometer" in Excel column AC. So, create a new column in your data called "Boost" and use the following equation:
"=14.696/101.3*(D#-AC#)" where # is the Excel row number
Open "File=>Option=>Columns and Profiles" in Virtual Dyno and put "Boost" into the "Boost" option. I had expected that by calling the column "Boost," Virtual Dyno would know what to do with it, but it didn't work out that way. ?? Make sure the check mark is on next to "Boost."
6. I actually prefer plotting MAP in KPa instead of boost. This makes it easy for me to correlate to my spark and fuel maps. To make this happen, open "File=>Option=>Columns and Profiles" in Virtual Dyno and put "MAP" into the "Boost" option field. Done. Couldn't be easier. Again, make sure the check mark is on next to "Boost."
7. There is also a TPS field in Virtual Dyno. In MS1, the thottle position value is called "TP" in Excel column E. I put "TP" into the "TPS" option field. Doesn't seem to have any effect. ??
Hints:
This is no different than a real dyno. To make valid comparisons, you need to eliminate variables so that, hopefully, only the effects of the tuning parameter you are working on are seen. To that end:
1. Do your comparison runs at the same time and place if possible. This is easy if you are just changing ECU parameters (make a run, reprogram ECU, make another run). Harder if you are changing mechanical items. If changing mechanical items, then use the same stretch of road and try to do compared runs under nearly identical weather conditions.
2. Be cognizant of aerodynamics. Windows, top and, for NAs, headlights will change your results. To get a "close" horsepower reading using the Mazda-published drag coefficient, you need to have windows closed, top up (or hardtop on), and NA headlights retracted.
3. Do two runs on the same stretch of road going opposite directions. This helps cut down on the effects of wind and slope. Your comparison plot would be the average of the two runs.
Bottom Line:
Does it work? Heck, yes, it works. Last night, I tried it in earnest for the first time. I wanted to compare two spark maps. The first is based upon the MSPNP default spark map as follows:
The second has a bit more timing, especially at high RPM:
Here's the result plotted by Virtual Dyno:
The additional timing at high RPM is clearly seen as increased torque and HP in that region. MUCH better than a butt dyno and, so far as I can tell, the data seems to be as useful as that from a real dyno. Now, if only I could hold a constant RPM . . . . Nonetheless, this looks like a valuable tuning tool to me.
Now, about the numbers . . . not too impressive. Keep in mind that my headlights were up. I'm certain that if they had been down, I would have gotten 270RWHP+ out of my non-intercooled Greddy.
Possible Bugs for Brad:
1. Need to specify AFR custom column name even if column name is "AFR."
2. Need to specify Boost custom column name even if column name is "Boost."
3. What does TPS do?
4. The auto update of car profiles overwrote a custom profile that I had stored. This was mildly annoying. It would be nice if there were a message that informs the user that new car profiles have been written.
5. I would love to be able to plot an average of multiple runs. That way, I could combine going two directions on a stretch of road. Doesn't seem like it would be too hard.
6. Help leaves a bit to be desired (there isn't any). Feel free to use anything I've written here to get that started.
7. Cool app. Glad to have you aboard. I'm not sure whether you are a Miata guy, but you owe it to yourself to take a drive one. If you're ever in the DFW area . . . .
8. You get a prop!!
Last edited by hornetball; 01-05-2012 at 03:29 PM.
Reply
1
1
01-06-2012, 07:56 PM
#28
Junior Member
Join Date: Jan 2012
Location: Kansas City, KS
Posts: 66
Total Cats: 3
reply to hornetball
This is what I've discovered so far playing around with Virtual Dyno:
1. Virtual Dyno uses the names found in the first row of a comma-delimited text file to get its parameters. The first row limitation is important. For example, a Megasquirt log writes the software version number on the first row and the parameter names on the second row. Therefore, you need to open that Megasquirt log and delete the first row for it to work with Virtual Dyno.
You are almost correct. If I have the logic built in the code it can trim headers for some types. HP Tuners has headers and I trim theirs out so this is just a matter of getting it supported. I know MS has several different versions 1,2 and 3. Until 1.1.6 drops only MS3 logs are supported but Im supporting MS2 logs as well with 1.1.6 and if someone has an MS1 log Ill try to get it supported as well. If I remember correctly you can drag and drop MS logs directly on 1.1.6 like all CSV and TXT files when it releases!!
2. Virtual Dyno names can be assigned to match what is in your file with a pulldown menu (under File=>Options=>Columns and Profiles). The parameters that Virtual Dyno needs to work its magic are:
-Time (Required for dyno plot)
-RPM (Required for dyno plot)
-AFR (Optional, needed if you want the AFR trace below the dyno plot)
-Boost (Optional, needed if you want the Boost trace below the dyno plot)
-TPS (This is listed in the menu, but I can't discern how/if it is used)
3. Because of #2 above, I really wouldn't pay much attention to the "list of supported ECUs" on Virtual Dyno's website. If you are handy with a sheet or text editor program and your ECU's output is human-readable text, you'll be able to use Virtual Dyno.
EXACTLY!!! Even unsupported logs can be graphed. Finally someone that understands.
MS1/MSPNP Instructions:
Here's an example of setting up with a log file from a MSPNP9093 (basically a Megasquirt I):
1. Open log file in Excel and delete first row. New first row should be parameter names.
2. Edit the data so that the pull you want to show is the only data present. This isn't strictly necessary, but you will get a lot of unwanted artifacts on your plot if you don't do this.
3. MS1 plots have "Time" and "RPM" columns. These are used as-is by Virtual Dyno.
4. MS1 plots have the wideband readout in volts under "O2" in Excel column F. To convert this to AFR, you need to know what the volts represent. In my case, I'm setup so that 0VDC = 10:1AFR and 5VDC = 20.8:1AFR. So, I created a new column in my data called "AFR" and used the following equation:
"=10+F#/5*(20.8-10)" where # is the Excel row number
I then opened "File=>Option=>Columns and Profiles" in Virtual Dyno and put "AFR" into the "AFR" option. I had expected that by calling the column "AFR," Virtual Dyno would know what to do with it, but it didn't work out that way. ?? Make sure the check mark is on next to "AFR."
5. If you want Boost plotted in units of PSI, then you need another calculated column. MS1 plots have a manifold pressure readout in KPa under "MAP" in Excel column D. MS1 plots also have an atmospheric pressure readout in KPa under "barometer" in Excel column AC. So, create a new column in your data called "Boost" and use the following equation:
"=14.696/101.3*(D#-AC#)" where # is the Excel row number
Open "File=>Option=>Columns and Profiles" in Virtual Dyno and put "Boost" into the "Boost" option. I had expected that by calling the column "Boost," Virtual Dyno would know what to do with it, but it didn't work out that way. ?? Make sure the check mark is on next to "Boost."
6. I actually prefer plotting MAP in KPa instead of boost. This makes it easy for me to correlate to my spark and fuel maps. To make this happen, open "File=>Option=>Columns and Profiles" in Virtual Dyno and put "MAP" into the "Boost" option field. Done. Couldn't be easier. Again, make sure the check mark is on next to "Boost."
7. There is also a TPS field in Virtual Dyno. In MS1, the thottle position value is called "TP" in Excel column E. I put "TP" into the "TPS" option field. Doesn't seem to have any effect. ??
Hints:
This is no different than a real dyno. To make valid comparisons, you need to eliminate variables so that, hopefully, only the effects of the tuning parameter you are working on are seen. To that end:
1. Do your comparison runs at the same time and place if possible. This is easy if you are just changing ECU parameters (make a run, reprogram ECU, make another run). Harder if you are changing mechanical items. If changing mechanical items, then use the same stretch of road and try to do compared runs under nearly identical weather conditions.
2. Be cognizant of aerodynamics. Windows, top and, for NAs, headlights will change your results. To get a "close" horsepower reading using the Mazda-published drag coefficient, you need to have windows closed, top up (or hardtop on), and NA headlights retracted.
3. Do two runs on the same stretch of road going opposite directions. This helps cut down on the effects of wind and slope. Your comparison plot would be the average of the two runs.
Bottom Line:
Does it work? Heck, yes, it works. Last night, I tried it in earnest for the first time. I wanted to compare two spark maps. The first is based upon the MSPNP default spark map as follows:
The second has a bit more timing, especially at high RPM:
Here's the result plotted by Virtual Dyno:
The additional timing at high RPM is clearly seen as increased torque and HP in that region. MUCH better than a butt dyno and, so far as I can tell, the data seems to be as useful as that from a real dyno. Now, if only I could hold a constant RPM . . . . Nonetheless, this looks like a valuable tuning tool to me.
Now, about the numbers . . . not too impressive. Keep in mind that my headlights were up. I'm certain that if they had been down, I would have gotten 270RWHP+ out of my non-intercooled Greddy.
Possible Bugs for Brad:
1. Need to specify AFR custom column name even if column name is "AFR."
Even though the column is named AFR I have no idea thats AFR as I try to detect the log type based on column names. If that particular log is supported then Ill be able to graph that column. Just because it says AFR doesnt mean the data is in a usable format. (ex. voltage)
2. Need to specify Boost custom column name even if column name is "Boost."
Same
3. What does TPS do?
TPS is used only to trim the junk out of the log that isnt part of the WOT pull. This trimming has a bug in 1.1.5 and 1.1.6 should fix it where that trimming cant be turned off by the option on the options screen.
4. The auto update of car profiles overwrote a custom profile that I had stored. This was mildly annoying. It would be nice if there were a message that informs the user that new car profiles have been written.
Im working on this. Im gonna address it when I release the car editor. Hang in there.
5. I would love to be able to plot an average of multiple runs. That way, I could combine going two directions on a stretch of road. Doesn't seem like it would be too hard.
Its hard to take an average unless datapoints are at the same RPM. I havent figure this request out either. It would give you a graph but it would be totally bogus.
6. Help leaves a bit to be desired (there isn't any). Feel free to use anything I've written here to get that started.
I would write a help file but I just don't have the time to do it. I really wish I could. It would be the best addition ever. If someone knows how to make a .chm file and wants to Ill add your name to the credits and publish it with the program. I could do the help file but that's time away from coding new features and fixing bugs.
7. Cool app. Glad to have you aboard. I'm not sure whether you are a Miata guy, but you owe it to yourself to take a drive one. If you're ever in the DFW area . . . .
Wish I had known this I was in DFW for the whole month of November and BORED. My new job had me there for training in a new programming language that's proprietary to their company.
8. You get a prop!!
1. Virtual Dyno uses the names found in the first row of a comma-delimited text file to get its parameters. The first row limitation is important. For example, a Megasquirt log writes the software version number on the first row and the parameter names on the second row. Therefore, you need to open that Megasquirt log and delete the first row for it to work with Virtual Dyno.
You are almost correct. If I have the logic built in the code it can trim headers for some types. HP Tuners has headers and I trim theirs out so this is just a matter of getting it supported. I know MS has several different versions 1,2 and 3. Until 1.1.6 drops only MS3 logs are supported but Im supporting MS2 logs as well with 1.1.6 and if someone has an MS1 log Ill try to get it supported as well. If I remember correctly you can drag and drop MS logs directly on 1.1.6 like all CSV and TXT files when it releases!!
2. Virtual Dyno names can be assigned to match what is in your file with a pulldown menu (under File=>Options=>Columns and Profiles). The parameters that Virtual Dyno needs to work its magic are:
-Time (Required for dyno plot)
-RPM (Required for dyno plot)
-AFR (Optional, needed if you want the AFR trace below the dyno plot)
-Boost (Optional, needed if you want the Boost trace below the dyno plot)
-TPS (This is listed in the menu, but I can't discern how/if it is used)
3. Because of #2 above, I really wouldn't pay much attention to the "list of supported ECUs" on Virtual Dyno's website. If you are handy with a sheet or text editor program and your ECU's output is human-readable text, you'll be able to use Virtual Dyno.
EXACTLY!!! Even unsupported logs can be graphed. Finally someone that understands.
MS1/MSPNP Instructions:
Here's an example of setting up with a log file from a MSPNP9093 (basically a Megasquirt I):
1. Open log file in Excel and delete first row. New first row should be parameter names.
2. Edit the data so that the pull you want to show is the only data present. This isn't strictly necessary, but you will get a lot of unwanted artifacts on your plot if you don't do this.
3. MS1 plots have "Time" and "RPM" columns. These are used as-is by Virtual Dyno.
4. MS1 plots have the wideband readout in volts under "O2" in Excel column F. To convert this to AFR, you need to know what the volts represent. In my case, I'm setup so that 0VDC = 10:1AFR and 5VDC = 20.8:1AFR. So, I created a new column in my data called "AFR" and used the following equation:
"=10+F#/5*(20.8-10)" where # is the Excel row number
I then opened "File=>Option=>Columns and Profiles" in Virtual Dyno and put "AFR" into the "AFR" option. I had expected that by calling the column "AFR," Virtual Dyno would know what to do with it, but it didn't work out that way. ?? Make sure the check mark is on next to "AFR."
5. If you want Boost plotted in units of PSI, then you need another calculated column. MS1 plots have a manifold pressure readout in KPa under "MAP" in Excel column D. MS1 plots also have an atmospheric pressure readout in KPa under "barometer" in Excel column AC. So, create a new column in your data called "Boost" and use the following equation:
"=14.696/101.3*(D#-AC#)" where # is the Excel row number
Open "File=>Option=>Columns and Profiles" in Virtual Dyno and put "Boost" into the "Boost" option. I had expected that by calling the column "Boost," Virtual Dyno would know what to do with it, but it didn't work out that way. ?? Make sure the check mark is on next to "Boost."
6. I actually prefer plotting MAP in KPa instead of boost. This makes it easy for me to correlate to my spark and fuel maps. To make this happen, open "File=>Option=>Columns and Profiles" in Virtual Dyno and put "MAP" into the "Boost" option field. Done. Couldn't be easier. Again, make sure the check mark is on next to "Boost."
7. There is also a TPS field in Virtual Dyno. In MS1, the thottle position value is called "TP" in Excel column E. I put "TP" into the "TPS" option field. Doesn't seem to have any effect. ??
Hints:
This is no different than a real dyno. To make valid comparisons, you need to eliminate variables so that, hopefully, only the effects of the tuning parameter you are working on are seen. To that end:
1. Do your comparison runs at the same time and place if possible. This is easy if you are just changing ECU parameters (make a run, reprogram ECU, make another run). Harder if you are changing mechanical items. If changing mechanical items, then use the same stretch of road and try to do compared runs under nearly identical weather conditions.
2. Be cognizant of aerodynamics. Windows, top and, for NAs, headlights will change your results. To get a "close" horsepower reading using the Mazda-published drag coefficient, you need to have windows closed, top up (or hardtop on), and NA headlights retracted.
3. Do two runs on the same stretch of road going opposite directions. This helps cut down on the effects of wind and slope. Your comparison plot would be the average of the two runs.
Bottom Line:
Does it work? Heck, yes, it works. Last night, I tried it in earnest for the first time. I wanted to compare two spark maps. The first is based upon the MSPNP default spark map as follows:
The second has a bit more timing, especially at high RPM:
Here's the result plotted by Virtual Dyno:
The additional timing at high RPM is clearly seen as increased torque and HP in that region. MUCH better than a butt dyno and, so far as I can tell, the data seems to be as useful as that from a real dyno. Now, if only I could hold a constant RPM . . . . Nonetheless, this looks like a valuable tuning tool to me.
Now, about the numbers . . . not too impressive. Keep in mind that my headlights were up. I'm certain that if they had been down, I would have gotten 270RWHP+ out of my non-intercooled Greddy.
Possible Bugs for Brad:
1. Need to specify AFR custom column name even if column name is "AFR."
Even though the column is named AFR I have no idea thats AFR as I try to detect the log type based on column names. If that particular log is supported then Ill be able to graph that column. Just because it says AFR doesnt mean the data is in a usable format. (ex. voltage)
2. Need to specify Boost custom column name even if column name is "Boost."
Same
3. What does TPS do?
TPS is used only to trim the junk out of the log that isnt part of the WOT pull. This trimming has a bug in 1.1.5 and 1.1.6 should fix it where that trimming cant be turned off by the option on the options screen.
4. The auto update of car profiles overwrote a custom profile that I had stored. This was mildly annoying. It would be nice if there were a message that informs the user that new car profiles have been written.
Im working on this. Im gonna address it when I release the car editor. Hang in there.
5. I would love to be able to plot an average of multiple runs. That way, I could combine going two directions on a stretch of road. Doesn't seem like it would be too hard.
Its hard to take an average unless datapoints are at the same RPM. I havent figure this request out either. It would give you a graph but it would be totally bogus.
6. Help leaves a bit to be desired (there isn't any). Feel free to use anything I've written here to get that started.
I would write a help file but I just don't have the time to do it. I really wish I could. It would be the best addition ever. If someone knows how to make a .chm file and wants to Ill add your name to the credits and publish it with the program. I could do the help file but that's time away from coding new features and fixing bugs.
7. Cool app. Glad to have you aboard. I'm not sure whether you are a Miata guy, but you owe it to yourself to take a drive one. If you're ever in the DFW area . . . .
Wish I had known this I was in DFW for the whole month of November and BORED. My new job had me there for training in a new programming language that's proprietary to their company.
8. You get a prop!!
Last edited by bbarnhill; 01-06-2012 at 08:39 PM.
Reply
0
0
01-06-2012, 08:13 PM
#29
Junior Member
Join Date: Jan 2012
Location: Kansas City, KS
Posts: 66
Total Cats: 3
Thanks,
Reply
0
0
01-06-2012, 09:51 PM
#30
For Computer Geeks
Brad,
Understand about "AFR" and "Boost." I just thought it might work the same as "RPM" and "Time."
MS1 log attached.
In case you haven't noticed, there is some serious brain power (coupled with irreverance) on miataturbo.net. I would say that JasonC SBB is one of the top minds around here when it comes to tuning. He is a particular fan of a derived parameter that divides Torque by MAP for an indication of engine efficiency. Check out this thread:
https://www.miataturbo.net/engine-performance-56/tale-3-turbo-setups-60618/
What do you think of an option to display this derived parameter? On the plus side, it is a very effective tool and would be simple to implement. On the minus side, few people will know about it or how to use it. ??
Also understand about the data being referenced to RPM points interfering with an averaging function. Typically, when I run into this situation (i.e., two different sets of data representing the same thing but containing slightly different index references), I'll use an interpolation algorithm so I can get the index references matched. Here's one that might work for you (in C because I'm a dinosaur):
double interpolate( double *rpm, double *val, int count, double rpm_idx )
{
// *rpm is a pointer to an array of RPM indexes in ascending order
// *val is a pointer to an array of values referenced to *rpm
// count is the number of valid values in *rpm and *val
// rpm_idx is the RPM value for which an interpolated value is desired
// This function returns the interpolated value from *val that
// corresponds to rpm_idx.
double ratio;
int idx = 1;
// Apply limits
if ( rpm_idx < rpm[ 0] ) rpm_idx = rpm[ 0];
if ( rpm_idx > rpm[count] ) rpm_idx = rpm[count];
// Get index position
while( rpm_idx > rpm[idx] )
{
idx++;
if ( idx == count ) break;
}
// Interpolate
ratio = rpm_idx - rpm[idx - 1];
ratio /= rpm[idx] - rpm[idx - 1];
return( val[idx - 1] + ratio * ( val[idx] - val[idx - 1] ) );
}
Understand about "AFR" and "Boost." I just thought it might work the same as "RPM" and "Time."
MS1 log attached.
In case you haven't noticed, there is some serious brain power (coupled with irreverance) on miataturbo.net. I would say that JasonC SBB is one of the top minds around here when it comes to tuning. He is a particular fan of a derived parameter that divides Torque by MAP for an indication of engine efficiency. Check out this thread:
https://www.miataturbo.net/engine-performance-56/tale-3-turbo-setups-60618/
What do you think of an option to display this derived parameter? On the plus side, it is a very effective tool and would be simple to implement. On the minus side, few people will know about it or how to use it. ??
Also understand about the data being referenced to RPM points interfering with an averaging function. Typically, when I run into this situation (i.e., two different sets of data representing the same thing but containing slightly different index references), I'll use an interpolation algorithm so I can get the index references matched. Here's one that might work for you (in C because I'm a dinosaur):
double interpolate( double *rpm, double *val, int count, double rpm_idx )
{
// *rpm is a pointer to an array of RPM indexes in ascending order
// *val is a pointer to an array of values referenced to *rpm
// count is the number of valid values in *rpm and *val
// rpm_idx is the RPM value for which an interpolated value is desired
// This function returns the interpolated value from *val that
// corresponds to rpm_idx.
double ratio;
int idx = 1;
// Apply limits
if ( rpm_idx < rpm[ 0] ) rpm_idx = rpm[ 0];
if ( rpm_idx > rpm[count] ) rpm_idx = rpm[count];
// Get index position
while( rpm_idx > rpm[idx] )
{
idx++;
if ( idx == count ) break;
}
// Interpolate
ratio = rpm_idx - rpm[idx - 1];
ratio /= rpm[idx] - rpm[idx - 1];
return( val[idx - 1] + ratio * ( val[idx] - val[idx - 1] ) );
}
Reply
0
0
01-07-2012, 09:32 AM
#31
Junior Member
Join Date: Jan 2012
Location: Kansas City, KS
Posts: 66
Total Cats: 3
Re: Virtual Dyno
Brad,
Understand about "AFR" and "Boost." I just thought it might work the same as "RPM" and "Time."
All columns graphed are treated exactly the same. The column definitions may have been there for those two (RPM and TIME) and not AFR and Boost for that particular logger.
MS1 log attached.
In case you haven't noticed, there is some serious brain power (coupled with irreverance) on miataturbo.net. I would say that JasonC SBB is one of the top minds around here when it comes to tuning. He is a particular fan of a derived parameter that divides Torque by MAP for an indication of engine efficiency. Check out this thread:
https://www.miataturbo.net/showthread.php?t=60618
What do you think of an option to display this derived parameter? On the plus side, it is a very effective tool and would be simple to implement. On the minus side, few people will know about it or how to use it. ??
I think with as many loggers that are supported by Virtual Dyno ... derived parameters are a scarey road for me to venture down. I will let you know that I have the functionality coded to graph any column in a log, Im just debating a paid version because donations are almost NON-existant.
Also understand about the data being referenced to RPM points interfering with an averaging function. Typically, when I run into this situation (i.e., two different sets of data representing the same thing but containing slightly different index references), I'll use an interpolation algorithm so I can get the index references matched. Here's one that might work for you (in C because I'm a dinosaur):
C is perfectly ok. VD is written in C# which is a C style language. Its easy to read any C style language. VB on the other hand ... *puke* Im going to have to look into this function and see if it will work for me.
double interpolate( double *rpm, double *val, int count, double rpm_idx )
{
// *rpm is a pointer to an array of RPM indexes in ascending order
// *val is a pointer to an array of values referenced to *rpm
// count is the number of valid values in *rpm and *val
// rpm_idx is the RPM value for which an interpolated value is desired
// This function returns the interpolated value from *val that
// corresponds to rpm_idx.
double ratio;
int idx = 1;
// Apply limits
if ( rpm_idx < rpm[ 0] ) rpm_idx = rpm[ 0];
if ( rpm_idx > rpm[count] ) rpm_idx = rpm[count];
// Get index position
while( rpm_idx > rpm[idx] )
{
idx++;
if ( idx == count ) break;
}
// Interpolate
ratio = rpm_idx - rpm[idx - 1];
ratio /= rpm[idx] - rpm[idx - 1];
return( val[idx - 1] + ratio * ( val[idx] - val[idx - 1] ) );
}
Understand about "AFR" and "Boost." I just thought it might work the same as "RPM" and "Time."
All columns graphed are treated exactly the same. The column definitions may have been there for those two (RPM and TIME) and not AFR and Boost for that particular logger.
MS1 log attached.
In case you haven't noticed, there is some serious brain power (coupled with irreverance) on miataturbo.net. I would say that JasonC SBB is one of the top minds around here when it comes to tuning. He is a particular fan of a derived parameter that divides Torque by MAP for an indication of engine efficiency. Check out this thread:
https://www.miataturbo.net/showthread.php?t=60618
What do you think of an option to display this derived parameter? On the plus side, it is a very effective tool and would be simple to implement. On the minus side, few people will know about it or how to use it. ??
I think with as many loggers that are supported by Virtual Dyno ... derived parameters are a scarey road for me to venture down. I will let you know that I have the functionality coded to graph any column in a log, Im just debating a paid version because donations are almost NON-existant.
Also understand about the data being referenced to RPM points interfering with an averaging function. Typically, when I run into this situation (i.e., two different sets of data representing the same thing but containing slightly different index references), I'll use an interpolation algorithm so I can get the index references matched. Here's one that might work for you (in C because I'm a dinosaur):
C is perfectly ok. VD is written in C# which is a C style language. Its easy to read any C style language. VB on the other hand ... *puke* Im going to have to look into this function and see if it will work for me.
double interpolate( double *rpm, double *val, int count, double rpm_idx )
{
// *rpm is a pointer to an array of RPM indexes in ascending order
// *val is a pointer to an array of values referenced to *rpm
// count is the number of valid values in *rpm and *val
// rpm_idx is the RPM value for which an interpolated value is desired
// This function returns the interpolated value from *val that
// corresponds to rpm_idx.
double ratio;
int idx = 1;
// Apply limits
if ( rpm_idx < rpm[ 0] ) rpm_idx = rpm[ 0];
if ( rpm_idx > rpm[count] ) rpm_idx = rpm[count];
// Get index position
while( rpm_idx > rpm[idx] )
{
idx++;
if ( idx == count ) break;
}
// Interpolate
ratio = rpm_idx - rpm[idx - 1];
ratio /= rpm[idx] - rpm[idx - 1];
return( val[idx - 1] + ratio * ( val[idx] - val[idx - 1] ) );
}
Reply
0
0
01-07-2012, 11:25 AM
#32
In any case, the curve I'm after is Torque (a derived parameter from Virtual Dyno) divided by Boost/MAP (an existing data set that Virtual Dyno uses).
The ability to graph any column in a log (which I do all the time in Excel) doesn't help because I need the internal derived Torque parameter from Virtual Dyno. Now, if I could save the derived HP and Torque data points in a text file . . . .
Reply
0
0
01-07-2012, 08:04 PM
#34
Got a log but my logs have very little detail...low resolution...dont know if its from the ecu or the laptop...
I'm using the 1.8 NA numbers with final drive ratio of the 10th AE per my source...2nd gear.
I'm guessing weight is about 2400lb... 100lb in upgrades seem too much for turbo and hardware? Full tank of gas,ps,ac,pw.
http://www.motortrend.com/roadtests/...a/viewall.html
What is it looking for as far as the boost and afr goes? i've tried using MAP, i've tried renaming it to boost and still doesnt show. Neither does AFR.
my log is edited so it works... the original unedited adaptronic log is the 5-4...
I'm using the 1.8 NA numbers with final drive ratio of the 10th AE per my source...2nd gear.
I'm guessing weight is about 2400lb... 100lb in upgrades seem too much for turbo and hardware? Full tank of gas,ps,ac,pw.
http://www.motortrend.com/roadtests/...a/viewall.html
What is it looking for as far as the boost and afr goes? i've tried using MAP, i've tried renaming it to boost and still doesnt show. Neither does AFR.
my log is edited so it works... the original unedited adaptronic log is the 5-4...
Last edited by triple88a; 01-07-2012 at 09:48 PM.
Reply
0
0
01-07-2012, 10:28 PM
#35
According to this website the drag for the NB is .37 and the frontal is 22.3
www.thinkthelen.com/mazda/miata.shtml
So far i have this...also weight is stock not with stuff added to it.
<?xml version="1.0" standalone="yes"?>
<CarDataset xmlns="http://tempuri.org/CarDataset.xsd">
<CarParameters>
<CarMake>Mazda</CarMake>
<CarModel>Miata 10th AE (6 speed)</CarModel>
<CarSubModel>1.8</CarSubModel>
<StartYear>1999</StartYear>
<EndYear>2000</EndYear>
<Weight>2300</Weight>
<TransTypeManual>true</TransTypeManual>
<TransGear2>2.270</TransGear2>
<TransGear3>1.650</TransGear3>
<TransGear4>1.260</TransGear4>
<FinalGearRatio>3.91</FinalGearRatio>
<DragCoefficient>0.37</DragCoefficient>
<FrontalArea>22.3</FrontalArea>
<TireDiameter>23.01</TireDiameter>
</CarParameters>
</CarDataset>
www.thinkthelen.com/mazda/miata.shtml
So far i have this...also weight is stock not with stuff added to it.
<?xml version="1.0" standalone="yes"?>
<CarDataset xmlns="http://tempuri.org/CarDataset.xsd">
<CarParameters>
<CarMake>Mazda</CarMake>
<CarModel>Miata 10th AE (6 speed)</CarModel>
<CarSubModel>1.8</CarSubModel>
<StartYear>1999</StartYear>
<EndYear>2000</EndYear>
<Weight>2300</Weight>
<TransTypeManual>true</TransTypeManual>
<TransGear2>2.270</TransGear2>
<TransGear3>1.650</TransGear3>
<TransGear4>1.260</TransGear4>
<FinalGearRatio>3.91</FinalGearRatio>
<DragCoefficient>0.37</DragCoefficient>
<FrontalArea>22.3</FrontalArea>
<TireDiameter>23.01</TireDiameter>
</CarParameters>
</CarDataset>
Reply
0
0
01-08-2012, 12:15 AM
#36
According to this website the drag for the NB is .37 and the frontal is 22.3
www.thinkthelen.com/mazda/miata.shtml
www.thinkthelen.com/mazda/miata.shtml
Well, it will give you some great HP readings.
Reply
0
0
01-08-2012, 12:37 AM
#37
The other thing I notice about your data is that it is irregular. Sometimes the rate speeds up to about 2Hz. So, it could be the ECU or laptop. Are you using Windows? Are there a lot of background tasks running?
The more data you can log the better. I'm afraid that your logging is far too slow to get anything useful right now.
I've found that when you make changes to the options, it may or may not automatically update the run you are displaying. You may need to reload the run to see the effect of the changed options.
I think your first order of business needs to be to find out why your data logging is so slow.
Reply
0
0
