261 M-Tronic - What makes it tick.....

[afleetcommand]

Does it work the same as Husqvarna "Autotunes"? The "Lean out tests" & feed back logic? Bet they do. Going to find out as a 441 is now a project saw. Acts a lot like the gen one autotunes....

I suppose I could layout that process (Husqvarna's) , but not interested in offending by insulting anyones intelligence. Most here probably already know, and its for a true Husqvarna type to do that amyway. Actually simple and really clever. As Mr. Snelling pointed out a few pages back, saws are simple in their requirements as compared to other types of powered vehicles or equipment. Unless there is a quantum leap in controls technology for small two strokes, adding all kinds of hooks and controls might fall into the diminishing returns for the expense in my most humble opinion. Autotunes "smarts" really don't need to focus on anything except RPM or pulses from the ignition module, Throttle position, and how long the little solenoid controlled valve stays closed for both testing and then tuning. Really simple. Really effective. Oh yeah, temp & throttle position for startups along with controlling that little valve. Then once running goes into "tuning" logic. And amazingly they work!

So I'm going to watch and hopefully learn about those M-Tronic systems.

 

[nohoff]

M-Tronic has only RPMs and a switch for starting Position.
No throttle position like Autotune.

The MS 441 is a Mtronic 1 saw and has no updateable ECUs available.

 

[Lightning Performance]

Boy they killed those early 441's fast.

 

[nohoff]

I did some research. The M-Tronic uses a LIN-Bus

 

[MustangMike]

Was wondering when something new would be in this thread!

 

[RI Chevy]

Bump

 

[treesmith]

Watch the clutch needle bearing on 261, mine went and it took the oiler arm with it, looks to have worn the clutch drum where bearing sits too and I shut it off as soon as it sounded funky

 

[Wonkydonkey]

That's been a problem here, clutch drum bearing/crankshaft wearing.

 

[Jason Gouw]

This thread is so interesting to me that I signed up and this is my first post on this forum.

I have an MS661 and have been studying the Mtronic system quite a bit. I am intrigued by what @breese has been finding out on his 261.

I found it to be remarkable that the ECU is able to extrapolate so much information such as load, altitude, fuel quality, etc. with such little external peripherals. I am guessing that there is a temperature sensor and barometric sensor built into the ECU. What I am unable to figure out is how the system is able to figure out optimum timing characteristics. This typically requires inputs on load, throttle position, and knock sensor if we want to get fancy.

My guess as to the plausible outputs and inputs the Mtronic ECU is capable of:

Outputs:
1. Spark Timing
2. Fuel metering

Inputs
1. RPM
2. Temperature (though not temperature at the carb, head, or exhaust which would be more beneficial. temp at the coil doesn't seem like a good spot as it would sense a skewed temperature based on the ambient temp and heat that's been radiated from the nearby engine).
3. Barometric pressure ("altitude")
4. Choke/Start position switch.
5(?). Spark ion sensing. Many cars have this now and wondering if Mtronic is "smart enough" to use this. This method basically uses the spark plug as a knock and combustion pressure/efficiency sensor by measuring the ionization voltage across the gap after the spark. Variance in combustion pressure will change the ionization voltage required to cross the spark gap (different than spark voltage and not nearly as high to create a spark).

My last point above is what I've been pondering on - whether the Mtronic system utilizes this method. Without it, I am hard pressed to believe that Mtronic can optimize for varying loads, fuel, etc. It would be a marketing gimmick (like many things are) without point number 5 above.

 

[Jason Gouw]

Further thoughts. Mtronic might be able to "guess" load and use this fuzzy logic to optimize fuel metering and timing. Since in the world of chainsaws, you either go zero throttle or WOT, perhaps Mtronic assumes WOT if the RPM increases above a certain threshold assuming that the start/choke detection switch remains open. By this method, the ECU can make calculations based on the assumption of WOT and rpm to get an idea of load, then use its fuel map based on temperature, pressure, rpm/load to give the "optimum" timing and fueling characteristics.

This method, although far from perfect, might make due without the spark ion sensing I mentioned in my previous post. It would also explain a lot behind the calibration procedures of Mtronic 1.0/2.0 (those 5 uniform cuts to be made assumes WOT and a loaded condition) and the drastically different Mtronic v3.0 calibration procedure (which requires WOT).

I am not very familiar with Husky's but have seen that their autotunes have a throttle position sensor. Wonder if their systems require similar calibration procedure as Mtronic. With the TPS, one could assume a far better and more accurate optimization of the fuel metering and timing.

On a slightly different note, I do have a scope. I'll hook it up to my 661's diagnostics port (really just a tap to ground and Mtronic control wire) and record the waveform. Will try to take a video and post.

 

[RI Chevy]

Algorithms.

 

[RI Chevy]

Any updates that can be reported? Or is this project shelved?

 

[nohoff]

The ECU uses the ignition timing for fuel metering.
Set the ign timing earlier and the machine has too much fuel -> higher RPMs
Close the fuel solenoid for 20 crankshaft rotations -> higher RPMs , the engine is running too rich | RPMs are going down -> MOAR FUEL!

 

[ABarrick]

Any of you fellas run into an Mtronic 261 that will not four stroke at wot with no load?

The most recent 261c I have will rev to the moon without burbling when piss revving. If you hold it wot for a second, at times it will limit itself with ignition (sounds like any old saw with a limited rpm coil, definitely not fuel burble) but it sounds way lean all the time.

 

[RI Chevy]

Is it stock?

 

[Jason Gouw]

On a slightly different note, I do have a scope. I'll hook it up to my 661's diagnostics port (really just a tap to ground and Mtronic control wire) and record the waveform. Will try to take a video and post.

Following up on my last post. Here are two videos I took of my MS661 connected to a scope.

First run on scope:

Second run on scope recorder:

My initial thoughts is that the signal is not a communication protocol. I have read others mention LIN Bus but this is not the case when it's running. I believe when an MDG1 diagnostic tool is connected to a non-running saw, it can communicate over LIN Bus protocol to extract ECU information and history but while it's running, the ECU sends drive signals to the solenoid valve. Looking at some of the zoomed-out view on the scope, the solenoid drive signal seems to be in a series of pulses. The individual pulse width does not seem to change but the off-time between each series of pulses do seem to change - which is why my scope was having a bit of trouble finding the leading trigger edge of each series of pulses when I accelerate.

I can probably get more information by hooking up the Mtronic to my logic analyzer and record the entire run pulse for pulse but I think this test has answered the question on whether or not LIN Bus communication is being used - the verdict is no, it's not used during the operation of the chainsaw. I don't have an MDG1 so I cannot comment on the diagnostic mode.

 

[RI Chevy]

Cool.

 

[RI Chevy]

Just Checking with this thread and if there have been any new findings?
@breese

 

[afleetcommand]

Worked on a couple now, with the Stihl Service tool. Wonder why the thing has to be running? ( U may know what I'm talking about ). The interface wasn't as useful as the Husqvarna CST. BUT who know's times and tools change. Nice saws, same basic technology as what's on the Husqvarna's. I happen to like the 461's, angling to get one.

 

[RI Chevy]

I think the Husqvarna software provides more useful information through the computer interface than Stihl software does.