Expanded coolant temps for aircooled engines - help wanted!

[msoultan]

Hey Guys,
I biting off a big one here for my first code mod, but this would be huge for us aircooled folks so I'm gonna give it a go.

I am currently talking with Eric to find out what limits exist in the MT software regarding coolant temps. If there are none, I am planning on making a menu (or configurator) option to allow the user to select water-cooled or air-cooled and adjust all the MT limits accordingly. I should hopefully find out any MT limitations from Eric in the near future.

Next, I wanted to find out if there are any temperature limitation built into the MS2Extra code, and if so, where. I am looking to expand the range from -40 to +500. I created a custom vbscript that was able to generate a new CLTFACTOR.INC file and the temperature range for my head temp sensor just fits perfectly within the 1024 ADC values.

So, I have a few other questions, but I think this is a good place to stop for now to identify what limits exist within the code and then move from there. Any help here would be much appreciated!

Thanks!
Mike

[msoultan]

Well, it looks like I'm good on the MT side. From what Eric was saying, I should be able to upload custom temp tables to the MS using some new file thing that Lance is working on. Plus, Roger said that MT should also be able to handle any temp range so that's good. For the time being I'll probably just compile the cltfactor.inc file straight into the code until the MT stuff is set up and ready to go with the custom tables.

Next, I'm planning on going through the MS2Extra INI file to adjust the temp limits. I would like some guidance on how the developers would like to proceed on this. Would you like me to set up a Configurator option (like used for CODE_VARIANT, IDLE_CONTROLLER, etc) such that the user can select AIRCOOLED or WATERCOOLED and have the appropriate temp ranges used throughout MT, or should I just increase the temp ranges for everything across the board. I am planning on using -40 to +500 unless I hear otherwise. Although it's a bit more work, I will most likely set up an option to select the temp range as it seems the most logical.

I've looked through the MS2Extra code and so far I haven't found any temp limits. I'm hoping that the developers might chime in to confirm that there are in fact no temp limitations.

thanks!
Mike

[msoultan]

This is pretty neat!

I just compiled in my modified cltfactor.inc file into the MS2Extra code and uploaded the new firmware. I powered it up, twisted the coolant pot on the stim, and the temp went all the way up to 500!

Now I'm going to start messing with the inputs in the INI file and see if I can start setting values larger than 200. I'm making good progress here!

Mike

[UnaClocker]

Mine goes up to 215f.. Not much more, but it is over 200.. My fan doesn't come on till 210, so I'm right at the edge of the limit. Would be nice if it went over 215 even on a watercooled car. If my engine is going super critical, I'd like to be able to see it.

[msoultan]

Doesn't water temp go even hotter than that when it turns to steam? That would be helpful to see instead of it railing at 215.

Btw, you might want to try updating the coolant table through MT 2.25 as it is hard-coded to use -40 and +350 as its limits. That might give you that extra information you're looking for.

That doesn't, however, allow you to input values above the limitations of the INI file. Which brings up another question. When I go through the INI file, I can also change the limits for the watercooled people as well. What limits/trigger temps would you suggest for the high end?

Thanks,
Mike

[UnaClocker]

I think up to 250f is fair for watercooled. The coolant is a water+antifreeze mix, under 16psi of pressure, it can easily reach 250f in liquid form.

[msoultan]

Then you'd probably want even higher than that so that you know how far into the danger zone you really are ;)

Give the "Calibrate thermistor tables" option under Tools in MT a try and see if you get values higher than 215. I'd be curious to see how that works out for you. While it won't allow you to input values higher than the current limits in the INI file (most around 200), it should allow the MS to report values up to 350.

Mike

[rb26dett]

there was some behaviour in the bg code to do with coolant temps at the high extreme... i forget what it was, but i didnt like it. i wonder if i could track it down.

i'd like to see it read whatever is there

that way if you are logging when something fails, you can see just how bad it got...

btw, although some parts of the coolant mix at X psi and X % mix may be able to get to 250f whatever that is in C, any parts exposed to heat energy will be locally boiling, and thus hotspotting. in the process it will be pushing coolant out of the overflow to make room. not good. best that its kept lower if possible.

not sure if i explained myself well then i can have another go if you would like.

fred.

[msoultan]

there was some behaviour in the bg code to do with coolant temps at the high extreme... i forget what it was, but i didnt like it. i wonder if i could track it down.

Yeah, I'd be curious to know what they were doing. Did it have to do with displaying of the temp or some kinda trigger at high temps?

[rb26dett]

http://www.msefi.com/viewtopic.php?t=16572

found it, about 4 posts down lance says fault mode...

i dont know if anything of this is left in, but i still think its a bad idea.

fred.

[msoultan]

http://www.msefi.com/viewtopic.php?t=16572

found it, about 4 posts down lance says fault mode...

i dont know if anything of this is left in, but i still think its a bad idea.

fred.

Yeah, that information is still in the lookup tables. If you want to get around it, resend your temp table using megatune's thermister tool and it will increase the temp limit to 350...

Mike

[masterx81]

If i've not misunderstood, the mod to megatune, affects only megatune, enabling it to see more than the limits, but the ecu (and it's functions) can read in any case only at the same range as before...
So, if you mod mt, and in the ms set the fan on temp at 300, it didn't work... But mt write the correct temp.
I not know why they have used the °F for the variable in the code...
Using °C (without modifing too much) can extend a lot the temp range, say, from -20°C to 215°C, are much more than actual -40°C to 101°C.
My car runs normally at 92°C, so it's really easy to touch the 101 limit...
And any aircooled engine is well over that this limit...

[Minami Kotaro]

My CLT readout in MegaTune goes up to at least 264F and the gauge wasn't at full sweep, yet.

I'm not using a real CHT gauge, just a sensor mounted in the cooling air coming off my head.

[msoultan]

If i've not misunderstood, the mod to megatune, affects only megatune, enabling it to see more than the limits, but the ecu (and it's functions) can read in any case only at the same range as before...
So, if you mod mt, and in the ms set the fan on temp at 300, it didn't work... But mt write the correct temp.

Until I actually change the INI file, you won't be able to input anything higher than the current limits.

I not know why they have used the °F for the variable in the code...
Using °C (without modifing too much) can extend a lot the temp range, say, from -20°C to 215°C, are much more than actual -40°C to 101°C.
My car runs normally at 92°C, so it's really easy to touch the 101 limit...
And any aircooled engine is well over that this limit...

I'm not totally sure why they didn't initially cover a larger temp range, but I'm guessing it was just because they'd end up having to skip a whole lot of temps to cover from -40 to +500. However, I think it was more of an oversight than anything.

Mike

[masterx81]

Probably they have done this because °F is the american temp unit, and a byte in ram is 8 bit, so 255 combintions...
Little simpler in sw development, but using °C wasn't that complicated (if planned during development...), now, a little more complicated because all sw (mt, megatunix,etc) must be updated to take the advantage of a bigger temp range...

[msoultan]

My CLT readout in MegaTune goes up to at least 264F and the gauge wasn't at full sweep, yet.

I'm not using a real CHT gauge, just a sensor mounted in the cooling air coming off my head.

Yeah, this is exactly why I want to make this mod. The whole cooling air measurement method is not the best method for metering the engine temp - BTDT! I want to work with realistic number that mean something to me, and are also more representative of the actual temperature of the engine, not the ambient air around it...

[msoultan]

Ok, here is the game plan. Since MT (and possibly MS2) currently supports displaying of temps in the -40F to 350F range, this is what I am going to use as the standard for the WATERCOOLED configuration in my MS2Extra INI mods. For AIRCOOLED, I am going to use -40F to 500F.

If anyone thinks otherwise, please let me know.

Thanks,
Mike

[efahl]

If i've not misunderstood, the mod to megatune, affects only megatune, enabling it to see more than the limits, but the ecu (and it's functions) can read in any case only at the same range as before...

In the MS-I (908) processor, temperatures are handled internally in a 1-byte value, which means you are limited to 0-255. Al looked at this and said, "ok, that's fine, but I want more range on the low end" so he offset those values by 40 to mean -40 - 215 (F) internally. Thus, no matter what you do (short of rewriting the whole friggin' embedded code), the MS-I codes will only know about temps up to 215.

On the MS-II, where temperatures are stored in a 2-byte signed value (-32768 to 32767), Al used a fixed point interpretation with no offset, so these values are interpreted to represent -3276.8 to 3276.7 degrees and hence you have basically unlimited range internally.

MegaTune was written from the outset to have no limitations (internal values used for gauges and such are floating point doubles, +/-1E308), so you won't find any problems calculating or displaying, say, exhaust or combustion temperatures much less coolant temps.

Eric

[masterx81]

If i've not misunderstood, the mod to megatune, affects only megatune, enabling it to see more than the limits, but the ecu (and it's functions) can read in any case only at the same range as before...

In the MS-I (908) processor, temperatures are handled internally in a 1-byte value, which means you are limited to 0-255. Al looked at this and said, "ok, that's fine, but I want more range on the low end" so he offset those values by 40 to mean -40 - 215 (F) internally. Thus, no matter what you do (short of rewriting the whole friggin' embedded code), the MS-I codes will only know about temps up to 215.

On the MS-II, where temperatures are stored in a 2-byte signed value (-32768 to 32767), Al used a fixed point interpretation with no offset, so these values are interpreted to represent -3276.8 to 3276.7 degrees and hence you have basically unlimited range internally.

MegaTune was written from the outset to have no limitations (internal values used for gauges and such are floating point doubles, +/-1E308), so you won't find any problems calculating or displaying, say, exhaust or combustion temperatures much less coolant temps.

Eric

Ok, THIS is an answer
Thanks for the explaination!

[msoultan]

Ok.. scale's being bumped up to 600f. That's what is on the VDO head temp sender so I figure it's probably a good limit to use.

I'm going to be updating the INI file soon. Stay tuned...