Copilot Download available for Power FC users

[Tem]

Can anyone think of a use for analog inputs (0-5V range)?

External gauges.  s8) s8)  Exhaust temp, boost, whatever anyone might have. I think the PFC already pretty much covers all the stock stuff.

[Anonymous]

Only one problem: no boost for me.  So sad.   s:wink: :wink: s:wink:

I can see adding a boost gauge and exhaust temperature gauges for the turbo guys.  I'd need some data on what the inputs would be... to convert to whatever units.  Want to help with that?

I can also see displaying knock as a graph if I can get about 65 (or even 30 or so) samples of that a second.  boarderner asked me to do that a while back.  He wanted to compare the knock sensor to what FC Edit displays (what comes out of the Power FC).  It will be interesting.  The program may have trouble drawing fast enough... VB6 drawing (even using API drawing functions) just isn't that fast.  I don't know why.

The data (AIs) are being logged (if the Phidgets IF Kit is detected), but not displayed yet.  That will be in 2.0.7.

I'm working on 2.0.7 already and just got done moving everything on the I/O tab onto a new window (I/O Settings, accessed via Settings menu) because, yet again, I exceeded the max number of controls you can have on a VB form.  What a pain, but now I have all kinds of room on the I/O Settings form for displaying AIs and can do things like add menu items and controls on the main form, which was the one that got too bloated.  The I/O Settings tab will let you assign the wideband O2 analog in to a Phidgets IF Kit AI instead of a Datalogit AI.  Same thing with the Oil Temperature AI.  It's all coded, but I'm going to add a whole lot more before uploading a new version.

[Tem]

Only one problem: no boost for me.  So sad.   s:wink: :wink: s:wink:

No worries, that's fixable.  s;) s;)

I can see adding a boost gauge and exhaust temperature gauges for the turbo guys.  I'd need some data on what the inputs would be... to convert to whatever units.  Want to help with that?

My GReddy gauges have linear 0-5V outputs for these:
ex temp: 300-1200C
temp oil/water: 30-150C
boost: 1st third 736-0mmhg, final 2/3s 0-2bar
(so 1.666...V = 0 in the boost gauge)

300C=572F, 1200C=2192F, 30C=86F, 150C=302F, 2.0bar=29.0075475psi  s8) s8)

mmhg is also used in US units, right?

I can also see displaying knock as a graph if I can get about 65 (or even 30 or so) samples of that a second.  boarderner asked me to do that a while back.  He wanted to compare the knock sensor to what FC Edit displays (what comes out of the Power FC).  It will be interesting.  The program may have trouble drawing fast enough... VB6 drawing (even using API drawing functions) just isn't that fast.  I don't know why.

If you graph it, does it have to be drawn that fast? It's not like you can look at the display all the time anyway, especially when driving at revs where you can "expect" knock. I think even slower drawing would be fine for that.  s8) s8)

[kanujunkie]

Only one problem: no boost for me.  So sad.   s:wink: :wink: s:wink:

No worries, that's fixable.  s;) s;)  

all this work and no boost!, get it sorted!!!  s:wink: :wink: s:wink:  

mmhg is also used in US units, right?

correct

[Anonymous]

Thanks for all the gauge info, Tem.

Come to think of it, I think that knock signal is audio.  I'd have to do some math, like get the RMS value... whatever the Power FC does... and I don't know what that is.  No, I shouldn't get distracted by that.  I have to finish doing some things I've already started.  

I'm almost 50 and have mostly outgrown my Need For Speed.  All this work I do on my car and the computer hardware/software is just so I have something interesting to do.  That's vitally important to me.  If I get super bored in a few years I'll buy a beater and turn the Spyder into an earth-scorching, silver missile.

[Tem]

Come to think of it, I think that knock signal is audio.

I'm pretty sure it is. The knock sensor itself is basically just a microphone tuned to wanted frequency range. I bet ECU gets the sound signal and just converts it into a number, so if you catch it before the ECU, you should get audio.

Some use the knock sensor to connect headphones to it.  s8) s8)

[Anonymous]

 s:idea: :idea: s:idea:   Not just an exhaust gas temp display, but also a "map" (indexed like the other 20x20 maps...RPM and Load) of min/max/avg temps.  A friend of mine with a carburated car, a race car (bracket racing is what he does), told me that he tunes by EGT.  Maybe there could be an auto tune option to get as close as possible to the EGT map.

----

I have the boost gauge almost done.  It is just a modification of the Autometer-like temperature gauge ActiveX control that I already did.  I can make it so it won't work unless they have a Power FC Gadget instead of Datalogit.  I don't think it is a good idea too put many more valuable things in the Lite (free) version that works with Datalogit.  (More gauges would be very valuable, IMO.)  Or how about tying the Boost and EGT gauges to a Phidgets 8/8/8 IF Kit -or- Power FC Gadget (but not Datalogit)?

[Anonymous]

How's this look?    s:twisted: :twisted: s:twisted:

[aaronjb]

I see VB.. it burns! Kill it with fire!

Ahem.. sorry  s;) s;)

Gauge looks nice, though  s;) s;)

[Tem]

12psi <3

[Anonymous]

It is kind of silly to have it go up to 30PSI.  What would be better: make the range something the user can set.   I 'spose there should be a way to change the units, too.  Later.

[heathstimpson]

Nice job Kevin; you certainly know your stuff  s8) s8)

[Anonymous]

Download this if you have Google Earth (which is a free download).  Right-click and save target as, then double-click it and it should open with Google Earth.  http://www.mr-s.org/copilot/20070714_155910.kml

That's just a few seconds of a log I made this morning, exported to Google Earth's KML format.  I did not have the wideband or accelerometer plugged in, otherwise that data would be in the info balloons, too.

By the way, that starts out in the parking lot of Clark's Fish Camp, which I wrote about a couple of weeks ago in another topic.  s;) s;)

[Tem]

DMAN! That looks great.  s:D s:D  

Now I wish I could log my drive right now. We're on vacation, just driving around without any destination or hurry.  s8) s8)

[Anonymous]

Thanks, Tem.  I've got some other tricks up my sleeve.  I took a day off from work to do that and I think it was worth it.  Have a nice holiday.

[Anonymous]

I finally figured out why Copilot was reporting bad data received while it was auto tuning the fuel (base map) online.  Quite a while ago I totally redid the way that was done and the problem was a software timing issue I hadn't thought of.  Anyway, I don't think this really affects anyone because I don't think anyone else enables the Auto Tune Fuel Online feature.   I'm going to do some more testing and hopefully upload a new version (3.1) this week.

[aaronjb]

I've been known to try it  s;) s;)

But it never seemed to achieve the same as offline auto tuning, for me.. for reasons that I couldn't quite fathom.

[Anonymous]

It was due to writing to the PFC while data was still being received and didn't seem to have any ill effects other than reading bad data (which it recognized as being bad and didn't use).  Prior to this fix, it was writing OK and I got good results, which I would assume would be the same as if I had done it working online then offline.  It goes through the same code except for what it actually does when something should be changed.  Then it has to stop acquisition (waiting for any pending reads complete was part of the fix), write the data, wait for writes to complete, read the same thing back (waiting for these reads to complete was also part of the fix), and restart acquisition.  

There's also a fix for not being able to check the box that enables online auto tuning fuel.   That was due to some changes to support other platforms.   Nothing much else to speak of.  I tried to average two dynos and it got to be too complicated, so I took that code out.   You will see that things were moved a little on the Dynamometer window where I made room for controls having to do with averaging, but I made those controls invisible since they don't work.  I may finish coding that later.

Definitely nearing completion, as there is very little else I can think of adding (and also have time to work on).  I wish I had another great idea that wouldn't take a whole lot of time to implement.

[Anonymous]

Here it is:   m http://www.mr-s.org/copilot/Copilot_3_1.zip m

Fixed bug: logic of online auto tune enables

Fixed bug: data type conversions

Fixed bug: "Target AFR Edit causes change in" combo box wasn't being initialized based on Registry.

Enhancement:  Save Launch Control settings regardless of Replay or not Replay

Enhancement: Made "Buttons" menu (used with hand controller) higher contrast for better daylight operation.

Testing: Minor stuff related to elapsed time (trying to figure out what program sometimes aborts/bombs).

Fixed bug: Don't use injector scale factor (pubInjSF) when comparing InjMS to base map value.

Testing: Try a different video capture synch method.

REMOVED:  (wasn't finished coding): Added capability to average two dynos and save as a 3rd one.

REMOVED: (wasn't finished coding): Took Tuner String out of Protect form and data file

Fixed bug (I/O error after auto tune writing base map)

[Anonymous]

Nifty MPG-related things including an Economy display similar to the one in the Toyota Prius... only better  s:) s:)     m http://www.mr-s.org/copilot/Copilot_3_2_3.zip m

[Anonymous]

Boy, this topic is really outdated.   I thought I'd give a status report for Tem and others interested.

I've been working hard on Ease of Use improvements.  It will be quite a bit more work -- maybe another week -- before I'm ready to release the new version.   Basically, what I'm doing is making it easier to get started, tell what should be done, and turn auto tuning on.

1.  Adding an Advisor window that will appear automatically (or not, if you decide you don't want that to pop up) at times like:
- Loaded all files in your working folder (or opened a different Auto Tune folder)
- Read everything from the Power FC
- Opened (read) a map from a file

The Advisor function will examine the maps and look for things like:
- Inj map is all 1's
- Target AFR map hasn't been defined
- Target AFR map doesn't correlate with Inj map
- Current and New maps don't match (which is OK some of the time)
- Min and/or Max maps are too close or too far away from the map they relate to (like Base, Ign)

The idea is to have the program take a look at what I'd want to know if I were helping someone get started and then tell them what I'd do (advice which they are free to ignore).   I'm going to start out with some obvious checks and expand the list of checks in future versions.  If I do a good enough job, it should take the place of having a complete and up-to-date User's Guide.

2.  Makes it easier to enable Auto Tuning of fuel
- If O2 Feedback is enabled, it will disable it for you before starting Auto Tuning
- If you cannot Auto Tune for some reason (like the Target AFR map or a Min or Max map has not been defined), it will tell you why you cannot do it yet.

3. Allows just editing the Inj map for people who are used to doing that (when the Inj map is edited, the Target AFR map is automatically changed).

4. Eliminates some confusing controls.

5. When O2 Feedback is disabled, previously acquired data (AFR samples) in the section of the map where O2 Feedback applies (cruise and idle) is automatically cleared because these AFRs were based on feedback.

6.  Added support for 3SGT1PRO Power FCs.   That is the Pro version (has ignition cut and launch control) for 3S-GTE engines.

7.  (Possibly) Allow editing and writing maps anytime you are connected to the Power FC.  Right now, when working online and doing data acquisition, you have to stop acquisition before you can do edits/writes.   This one is very complicated to implement.  May save this for later.

and more...

[Tem]

Looking good Beanie!  s:D s:D  

I forgot there was a thread about this over here as well...

[custardavenger]

Cool. I'm really looking forward to this. Couldn't quite get my head round the old version enough to use if confidently.

Good work Beanie

[Anonymous]

I decided to take it a step further and implement a "Fix It" button.    There's a new "Intelligence" (not very intelligent, but it'll do) module that makes all kinds of checks as the program runs, basically checking everything I can think of, and creates Advisory Messages for a new window.   The Advisory Messages explain what the situation is and in some cases list steps to correct it.  All messages start out with an identifier like "A011" then a colon and the rest of the message.  Example:

A21: Target AFR map has not been defined so you cannot auto tune fuel injection

If someone has a problem, they can refer to it by this unique ID, not the whole message.  "A" is for Advisory.  The next character indicates severity ("0" is something that can be ignored, "1" prevents some feature from working (other than auto tuning fuel), "2" prevents auto tuning fuel, and other 2nd characters I don't have assigned yet).  Then comes the unique advisory number (1 in this case).   On the new window ("Advice") which displays these message, there will be a drop-down list of Advisory IDs like

A011
A12
A21

You can select one from the list and if it is something the program knows how to fix, a "Fix It" button will be enabled right next to the drop-down list.  If there is more than one way to fix it, it will give you choices, otherwise it will just fix the problem.   An example problem is that you have not defined the Target AFR map.   If your Inj map isn't all 1.0 (which is what it would be if someone did a Recalculate Base using FC Edit...  that's bad but can be fixed by a series of steps that you'll have to do yourself), it most likely represents the commanded AFRs, so the program will convert the Inj map values to Target AFR values.

Another example advisory:

A113: Gears have not been learned.  This prevents using the Dynamometer function.  To learn the gears (up to 6, depending on your transmission):
1. Drive for several seconds in each gear, holding the speed steady.
2. Observe ...  (more help steps will have to go here)

I don't know about you folks, but I think that's better than a manual and it is more fun for me to write the code than to add to the User's Guide.

The Fix It function means that this will take quite a bit more work.  It may be 2-3 weeks before I have it all coded and tested.

[custardavenger]

Sounds good Beanie. Can't wait to see what you come up with and give it a go.