[rusEfi] FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Your chance to introduce yourself and your vehicle
Post Reply
User avatar
sir
running engine in first post
running engine in first post
Posts: 9
Joined: Fri Dec 29, 2017 11:08 am
Location: Cracow, PL

FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by sir »

Hello from Poland!

I would like to present my EFI retrofit project based on rusEFI.

Engine: S-31 type (used in FSO/FSM Syrena model 104, 105, 105L, R20 and BOSTO) -- pretty similar to Wartburg 353 engine
Type: Water-cooled, three-cylinder 2-stroke, piston-controlled intake
Mixture control: float-type carburetor
Fuel: premix 1:30
Capacity: 842 ccm
Rated output: 40 PS at 4300 rpm
Max. torque: 80 Nn at 2750 rpm

Parts used:

- fuel pump: universal 12V 4bar
- fuel filter: with integrated 3bar pressure regulator
- crank sensor: 5WK9 607 (opel astra, vectra)
- CLT sensor: NTC 10k M8 termistor
- IAT sensor: NTC 10k termistor
- MAP sensor: 16212460 (np. Opel Corsa)
- full Triumph 955i ITB body with fuel rail, injectors, IAC stepper motor and TPS
- some laser-cut parts designed by my
- STM32F4 Discovery kit mounted on prorotype board (only ADC input buffer, stepper driver -- mostly based on Frankenso circuit and 6x irgb14c40l IGTB transistors for injectors/coils driver)

Images can be enlarged by opening in new browser tab

My FSM Syrena 105L, 1976 - this car is not well known in the world, mainly its distribution took place in Poland during PRL times


engine bay - currently I have test engine installed in self-made test bench, see next photos


intake manifold adapter - plywood prototype and final metal one


crank sensor adapter


36/1 trigger wheel


Test bench





Engine just before first tests


First successful start attempts videos (as it turned out later with a damaged MAP sensor):




Latest test (with working MAP sensor):



Map sensor is connected only to middle cylinder throttle and has optimized measuring window (between inlet window opening start and wide open) set in rusefi firmware. As you can see in the film, the vacuum pressure varies only between 73 and 100 kPa (general 'problem' of 2 stroke engines). This can be a problem because I want to try with Speed Density first -- this can be difficult to tune well. In case of failure I will try alpha-N. I must also deal with interferences from ignition circuit (mostly from high voltage part) because sometimes there are false signals collected from crank sensor input. I think Coil-On-Plug coils type can help with that. At the moment engine starts, works witch SD but requires further tuning.
User avatar
AndreyB
Site Admin
Posts: 14280
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by AndreyB »

Welcome to the forum and welcome to the elite "running engine in first post" group!

Ps: can you please attach your currenttune.msq file? I want to build a collection of those.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
Abricos
contributor
contributor
Posts: 849
Joined: Mon Aug 18, 2014 12:32 am
Location: Carteret, NJ 07008

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by Abricos »

WoW wonderful description ...
for 4 weeks several thousand video views ...
it's such a popular Engine ???
User avatar
AndreyB
Site Admin
Posts: 14280
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by AndreyB »

Abricos wrote:
Mon Jan 22, 2018 10:07 pm
for 4 weeks several thousand video views ...
it's such a popular Engine ???
I am a google ninja! http://www.syrena.nekla.pl/forum/viewtopic.php?t=23771
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
sir
running engine in first post
running engine in first post
Posts: 9
Joined: Fri Dec 29, 2017 11:08 am
Location: Cracow, PL

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by sir »

russian wrote:
Mon Jan 22, 2018 6:59 pm
Welcome to the forum and welcome to the elite "running engine in first post" group!
Thx, good that finally we managed to activate my account and I could send my first post ;)
russian wrote:
Mon Jan 22, 2018 8:18 pm
Ps: can you please attach your currenttune.msq file? I want to build a collection of those.
Sure, but I have to admit, that for now VE map is in pretty bad shape. Map was, let's say 'tuned by ear' ;) I plan to buy a wideband, but I'm not sure how long will it operate well due the unburned oil in exhaust gases. Any hints on this topic will be appreciated.

Ignition map is flat -- 22,5 deg BTDC. Oryginally this engine has no distributor, but one contact breaker, one capacitor and one ignition coil for each cylinder. Additionaly no ignition advance mechanism. 22,50 deg BTDC is the factory preset for best operation at max. torque rpm (2750). It is known that this can't work optimally for the remaining engine speed range, so some time ago I constructed a contact-less ignition module controlled by lpc1114 with programmed ignition advance based on rpm:



and I have some ignition maps to start with, but at first I want to tune VE map well with stock configuration to avoid impact of any unclear factors.
Abricos wrote:
Mon Jan 22, 2018 10:07 pm
WoW wonderful description ...
for 4 weeks several thousand video views ...
it's such a popular Engine ???
someone has put a link to my post on Syrena forum at the Polish equivalent of reddit - that's why videos have so many views.
russian wrote:
Mon Jan 22, 2018 10:23 pm
I am a google ninja! http://www.syrena.nekla.pl/forum/viewtopic.php?t=23771
exactly this link ;)
Attachments

[The extension msq has been deactivated and can no longer be displayed.]

puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by puff »

This lpc1114-based ignition - what was the signal source? was it the 36-1 wheel?
User avatar
sir
running engine in first post
running engine in first post
Posts: 9
Joined: Fri Dec 29, 2017 11:08 am
Location: Cracow, PL

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by sir »

puff wrote:
Mon Jan 22, 2018 11:10 pm
This lpc1114-based ignition - what was the signal source? was it the 36-1 wheel?
Three unipolar TLE4905 hall sensors (one for each cylinder) triggered by neodymium magnet in aluminium bracket mounted at original cam. Mainly because then I wanted to fit whole pcb inside housing of the oryginal ignition without visible additional elements outside.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by kb1gtt »

Yes, well done, and congrats on being part of that elite group. Welcome to the forum.

This is a link you might have interest in.
https://rusefi.com/forum/viewtopic.php?f=4&t=286

Also have you found the wiki? If not you might find this link of interest.
https://rusefi.com/wiki/index.php?title=Manual:Hardware#J701_NEON_engine_schematics
Welcome to the friendlier side of internet crazy :)
User avatar
AndreyB
Site Admin
Posts: 14280
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by AndreyB »

Can I please use this image for marketing purposes? It's just so.. beautiful!
Attachments
3_cylinder_diy.jpg
3_cylinder_diy.jpg (147.79 KiB) Viewed 14075 times
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
sir
running engine in first post
running engine in first post
Posts: 9
Joined: Fri Dec 29, 2017 11:08 am
Location: Cracow, PL

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by sir »

russian wrote:
Tue Jan 23, 2018 3:42 am
Can I please use this image for marketing purposes? It's just so.. beautiful!
Oh, yes please. I know it looks funny, but it was the fastest way to build a prototype. It took me only a few hours to build it. I wanted to start testing as soon as possible :lol: I plan to do my own PCB with small dimensions, but maybe at a later stage.
User avatar
sir
running engine in first post
running engine in first post
Posts: 9
Joined: Fri Dec 29, 2017 11:08 am
Location: Cracow, PL

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by sir »

kb1gtt wrote:
Tue Jan 23, 2018 1:59 am
This is a link you might have interest in.
https://rusefi.com/forum/viewtopic.php?f=4&t=286
Thx, I have a good experience with irgb14c40l from my other ignition projects. But I will look at this link too.
kb1gtt wrote:
Tue Jan 23, 2018 1:59 am
Also have you found the wiki? If not you might find this link of interest.
https://rusefi.com/wiki/index.php?title=Manual:Hardware#J701_NEON_engine_schematics
Yes, I know this wiki, I started playing with EFI from that, long time before I created an account on the forum.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by kb1gtt »

I don't know, I think he's got to many components :)

I like the simplicity. I also paused on that picture and looked at it for a while. Just like art, I looked at it upside down, and tried to let it speak to me. When I see that picture it says "Hey, I've got stuff to do, so lets get it done."
sir wrote:
Tue Jan 23, 2018 8:02 am
I plan to do my own PCB with small dimensions, but maybe at a later stage.
If you do something, I could offer some help. It looks like your setup is very simple, so it probably would not take to long. Do you think you might put it inside a distributor cap or something like that? If you design your own, if you post it somewhere I can review it and offer feedback. Or if you get me some of the details of how you want to mount it and such, I could likely draft something for you. Either that, or of course a Frankenso could do most of what you're doing, but with external igniters. If you don't ignore the cost of your labor, the Frankenso would likely be the lowest cost solution. Have you seen Prometheus? We have heard good things from a couple of people who have done that board layout. See the below link.
https://rusefi.com/forum/viewtopic.php?f=4&t=1215

Also I wonder how @ would feel about adding multiple map support. If you could get one MAP per cyl, for individual cyl control, would that be of interest? In the past I've made a board I called MultiMap, it combined multiple map sensors into one, reporting the lowest signal. I understand that the ITB folks commonly use a network of small hoses and small check valves to combine the signals. This did that same thing but electronically. Bu using the lowest MAP signal, the other cyl's tend to run a pinch rich, which is OK from a performance standpoint.
https://github.com/jharvey/MultiMAP

About WO2, I would suggest Exhaust Gas Temperature (EGT) sensors. Thermocouples in the exhaust are commonly used to indicate if you have an engine problem. WO2's are typically for emissions purposes. The entire WO2 tuning is kind of blah. If power is your concern, typically you would tune such that you are rich, then back it down little by little until you see peak power. then you use EGT's looking for abnormal temperatures which would indicate engine problems. The O2 feedback is really only helpful when you don't know if your fuel pump is delivering enough fuel. AKA you could command extra fuel, but it might not get delivered if your fuel rail pressure is dropping. The other place that the O2 feedback is handy is for the auto-tuning. It's kind of nice to let the computer get you close to peak HP with out the human having to type numbers into the table.
Welcome to the friendlier side of internet crazy :)
User avatar
sir
running engine in first post
running engine in first post
Posts: 9
Joined: Fri Dec 29, 2017 11:08 am
Location: Cracow, PL

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by sir »

kb1gtt wrote:
Tue Jan 23, 2018 10:59 am
sir wrote:
Tue Jan 23, 2018 8:02 am
I plan to do my own PCB with small dimensions, but maybe at a later stage.
If you do something, I could offer some help. It looks like your setup is very simple, so it probably would not take to long. Do you think you might put it inside a distributor cap or something like that?
Putting all into original ignition breakers cover might be difficult...



This is my previous ignition system conversion:

original ignition module with contact breakers plate looks like this:


my lpc1114-based board with variable ignition timing:


As you can see in the photos above, the stm discovery board will definitely not fit. Even without discovery board, placing the controller itself with all necessary peripherals on such a small PCB will be rather impossible. Until now, I wanted it to look like the original, but after conversion from the carburetor to the ITB EFI, it will be hardly possible to hide such a modification... so I do not need to fit everything in the original ignition housing, right? :)
kb1gtt wrote:
Tue Jan 23, 2018 10:59 am
If you design your own, if you post it somewhere I can review it and offer feedback. Or if you get me some of the details of how you want to mount it and such, I could likely draft something for you.
This can be very helpful, thx -- I'll keep that in mind.
kb1gtt wrote:
Tue Jan 23, 2018 10:59 am
Either that, or of course a Frankenso could do most of what you're doing, but with external igniters. If you don't ignore the cost of your labor, the Frankenso would likely be the lowest cost solution. Have you seen Prometheus? We have heard good things from a couple of people who have done that board layout. See the below link.
https://rusefi.com/forum/viewtopic.php?f=4&t=1215
I want to simplify the PCB and design in such a way that it would have only what I really need. That's why I do not want use Frankenso. I've seen Prometheus -- yes, it's smaller, but I think I still want to do my own :)
kb1gtt wrote:
Tue Jan 23, 2018 10:59 am
Also I wonder how @ would feel about adding multiple map support. If you could get one MAP per cyl, for individual cyl control, would that be of interest? In the past I've made a board I called MultiMap, it combined multiple map sensors into one, reporting the lowest signal. I understand that the ITB folks commonly use a network of small hoses and small check valves to combine the signals. This did that same thing but electronically. Bu using the lowest MAP signal, the other cyl's tend to run a pinch rich, which is OK from a performance standpoint.
https://github.com/jharvey/MultiMAP
About that... Currently I use modificated rusefi firmware to sample MAP sensor only once per rev (because i have MAP sensor connected to only one cylinder -- no common collector with the other 2 cylinders):

Code: Select all

diff --git a/firmware/controllers/map_averaging.cpp b/firmware/controllers/map_averaging.cpp
index 06495ab5..ff293d3c 100644                                                                                                                                                                 
--- a/firmware/controllers/map_averaging.cpp                                                                                                                                                    
+++ b/firmware/controllers/map_averaging.cpp                                                                                                                                                    
@@ -227,7 +227,7 @@ static void mapAveragingCallback(trigger_event_e ckpEventType,                                                                                                              
        measurementsPerRevolution = measurementsPerRevolutionCounter;
        measurementsPerRevolutionCounter = 0;
 
-       for (int i = 0; i < engineConfiguration->specs.cylindersCount; i++) {
+       for (int i = 0; i < 1; i++) {
                angle_t samplingStart = ENGINE(engineState.mapAveragingStart[i]);
 
                angle_t samplingDuration = ENGINE(engineState.mapAveragingDuration);
That (now I am not sure that such a change in the code is really necessary) connected with proper averaging start angle and duration angle gives a stable MAP reading form middle cylinder. Averaging 3 independent MAP sensors in my opinion will not increase the measurement range (currently 100-75 kPa) and the stability of readings.

Some info in subject of poor vacuum pressure in two-stroke engines and difficulties in tuning with MAP: http://www.msextra.com/forums/viewtopic.php?t=65104
kb1gtt wrote:
Tue Jan 23, 2018 10:59 am
About WO2, I would suggest Exhaust Gas Temperature (EGT) sensors. Thermocouples in the exhaust are commonly used to indicate if you have an engine problem. WO2's are typically for emissions purposes. The entire WO2 tuning is kind of blah. If power is your concern, typically you would tune such that you are rich, then back it down little by little until you see peak power. then you use EGT's looking for abnormal temperatures which would indicate engine problems. The O2 feedback is really only helpful when you don't know if your fuel pump is delivering enough fuel. AKA you could command extra fuel, but it might not get delivered if your fuel rail pressure is dropping. The other place that the O2 feedback is handy is for the auto-tuning. It's kind of nice to let the computer get you close to peak HP with out the human having to type numbers into the table.
This is a topic in which -- I must to admit -- I need to dig deeper. Even if I will use WO2 it will be used only for presetting/autotune, without any later operation in a closed loop (will be dismounted).
Last edited by sir on Wed Jan 24, 2018 8:10 am, edited 1 time in total.
User avatar
AndreyB
Site Admin
Posts: 14280
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by AndreyB »

sir wrote:
Tue Jan 23, 2018 11:36 pm
About that... Currently I use modificated rusefi firmware to sample MAP sensor only once per rev (because i have MAP sensor connected to only one cylinder -- no common collector with the other 2 cylinders):
it would be a tiny fix to put this into the official branch https://github.com/rusefi/rusefi/issues/549
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
sir
running engine in first post
running engine in first post
Posts: 9
Joined: Fri Dec 29, 2017 11:08 am
Location: Cracow, PL

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by sir »

russian wrote:
Wed Jan 24, 2018 2:12 am
sir wrote:
Tue Jan 23, 2018 11:36 pm
About that... Currently I use modificated rusefi firmware to sample MAP sensor only once per rev (because i have MAP sensor connected to only one cylinder -- no common collector with the other 2 cylinders):
it would be a tiny fix to put this into the official branch https://github.com/rusefi/rusefi/issues/549
It would be very nice, but universal logic might be a little tricky. In the 4-stroke engine the intake stroke is once every two crankshaft rotations, and can be measured in angle after TDC. In a two-stroke engine, the 'intake stroke' occurs once per crankshaft rotation and can be measured in the angle that occurs before TDC of each clinder.

For my engine it is between 55 and 0 degrees before TDC of each cylidner -- this is angle window when piston uncover inlet port (during up movement) and mixture can be succed into crankcase. After my modification the MAP measurement only occurs for the first cylinder, so the measuring window must be in reference to the TDC of the first cylinder. For a 3 cylinder engine, each cylinder has a 120 degree shift in phase...

let's do some math:
- inlet opening start for 2nd cylinder: 55 deg BTDC + 120 deg = 65 deg ATDC of first cylinder
- inlet full open: 0deg BTDC + 120 deg = 120 deg ATDC of first cylinder

So in this case averaging should start at 65 deg ATDC of first cylinder and averaging window should be 55 deg (120-65) -- this are exactly settigns you can find in CurrentTune.msq which I attached a few posts above. In current shape of firmware there is no way to do it differently -- you can't specify averaging start window begining before TDC and can't to choose a specific cylinder.

MAP sampling start angle option set to 305 deg (-55+360 which is start opening window for the next engine rev) could be a workaround for first issue. I do not know if current firmware can handle it properly -- but TunerStudio certainly not (can't set this value above 90 deg, which is generally ok for 4-stroke engines).
User avatar
AndreyB
Site Admin
Posts: 14280
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by AndreyB »

Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
sir
running engine in first post
running engine in first post
Posts: 9
Joined: Fri Dec 29, 2017 11:08 am
Location: Cracow, PL

Re: FSM Syrena 105L S-31 two stroke engine EFI conversion #41

Post by sir »

Wow, that was quick! I just compiled fresh firmware and flashed testboard with it. It's working :) Checked in engine sniffer only, but it seems to work just as expected. For MAP on 1st cylinder I can set start angle at 305 deg or -55 deg, for 2nd at 65 deg and for 3rd at 185 deg. That's perfect! :)
Post Reply