2010 RTV 500 repairs

[kmelander]

I hope you have replaced the spark plugs, which solves 85% of the poor running issues with the Subaru engine used in the RTV500.
We sell and lease a lot of RTV500s, they are very popular with farmers/ranchers, horse folks, landscapers, and housing communities. Several of the local colleges and hospitals have a couple or more.

It is pretty simple to do all of the components ohm tests and the harness ohms tests covering the harness and the ECU. The components are for the most part easily accessible.

There is a fuel filter on the inlet side of the fuel pump in the fuel tank. Remove the pump/sender assembly and wash the sock filter.

I did change the plugs with no luck. Then I changed the injector, also no improvement. Throwing parts at this thing is expensive. I asked the mechanics at my local dealer if the coils were known to partially fail and create this issue but they were less than encouraging. They did however mention the fuel pump specifically so, that was encouraging. I'm not a FI expert. Can a bad fuel pump, presumably causing low fuel pressure to the injector, cause a rich condition? Even if it were, it seems improbable that it would persistently misfire. Also, I don't think it's misfiring on either cylinder with prejudice, it's seems to be both cylinders misfiring. At least that's what the spark plugs are telling me. It really feels like an ignition issue to me. I'm really hoping its the coil, since it's a simple fix. An interesting conclusion - the test for the coil 2ndary is measuring resistance from plug cap to plug cap, which means this coil is not cylinder discriminate. It must fire both plugs every 360 deg. This means one cylinder is always firing at the top of the exhaust stroke? That seems...odd. I don't know, maybe it's common in the world of these 360 deg. parallel twins. As I've said previously, I've never worked on one of these engines, or any other engine configured like this.

I will say this, again, the guys at the dealership were great. They were sympathetic and encouraging. Very busy and stretched thin on time, but still talked with me. Ultimately offered to handle it for me if I get stumped, which was encouraging but made my wallet twitch nervously.

 

[lugbolt]

yep low fuel pressure can absolutely cause misfires. I'm dealing with a Polaris 850 at the moment (at home) which has a misfire, to "me" it's an obvious ignition misfire-but it has 42 psi fuel pressure. Should be 60. New fuel pump comes in today; and I'm thinking it'll solve the issue.

UPDATE: Yep-60 psi fuel pressure with the new pump, runs great.

 

[kmelander]

29.95k ohm on the 2ndary of the coil, 3.5 ohm on the primary. 12vdc to the pin. It's just barely out of spec on the high side of the secondary. I guess that means replace it?

 

[kmelander]

yep low fuel pressure can absolutely cause misfires. I'm dealing with a Polaris 850 at the moment (at home) which has a misfire, to "me" it's an obvious ignition misfire-but it has 42 psi fuel pressure. Should be 60. New fuel pump comes in today; and I'm thinking it'll solve the issue.

UPDATE: Yep-60 psi fuel pressure with the new pump, runs great.

That's great! I'm waiting for my new fuel pressure test kit. I did find my old one but it doesn't have any fittings to go to a barbed connection. I re-set the governor control linkage (per shop manual) and tested some sensors last night. The IA and ECT (temp) sensors checked out ok. I need to repeat the "Air pressure sensor" test. I measured resistance between the 3 pins on it but the manual does not give these values. Instead they are giving voltage specs. I sure hope the fuel pressure is the ah-ha moment for me too. It's a bit humiliating to be getting my butt kicked by this little wheezer engine. Something else that crossed my mind is electrical grounds. I imagine the ECU and associated electronics are sensitive to this and ground issues have a way of elusively creating problems. If the fuel pressure checks out ok, I'll add that to the list of seemingly random items to check.

One thing that's bugging me...my compression reading. The shop manual states 165#-185# as acceptable. I'm getting 200# on each cylinder and that feels odd to me but I can't create a situation in my head where an improperly timed/adjusted valvetrain would cause this, or could it? Another interesting point; the ecu isn't recognizing any errors (I assume) because it's not throwing any idiot lights on the dash. This could also be an indication that the ecu is bad I suppose. I've never wished for a bad fuel pump before, but I sure am now.

 

[kmelander]

I tested the fuel pressure and it was fine. It started off at about 34# with engine off and climbed to 38# with the engine running. Then I decided to test the crank sensor, which tested ok thankfully. Then I decided to test the intake air pressure sensor.

This was a bit confusing by the instructions. Hopefully I performed the test correctly. Specifically, I think step #10 is out of order. At least I can't figure out how to measure the resistance of the wire between the ECU connector and the sensor pin with the ECU plugged in and powered up. I do think I found a problem...with the ECU! Step #11 calls for a resistance reading between the center pin of the harness connector and body ground. I performed this test with the ECU plugged in and powered up. When the fuel pump was running, I was getting around 33 ohms. Once the fuel pump stopped, I was getting a constant bounce reading between 1.x and 3.x ohms. According to step #13 and the table at the bottom of the page, the ECU is bad? With the key off, I do get a reading of 0 ohms between body ground and the center pin of the sensor. All of the other tests passed.

 

[kmelander]

Update:
Based on the MAP sensor testing, I ordered and received a replacement ECU. I purchased it used from an Ebay seller. I installed the new ECU and performed the same MAP sensor test to see if anything was different, which to my disappointment, it was not. One demerit for the WSM on this. The center pin to ground resistance behaved the same way. I started and ran the engine which ran exactly the same as it did with the original ECU. The only bright spot was that the ECU cooling fan did run for a period after starting the engine, which did not happen with the old ECU. It may just be a software version difference. The model year this part was removed from was a 2009 and mine is a 2010.

The Ebay seller had some additional take-off parts available. Specifically, the MAP sensor and the ignition coil, which I purchased. I'm still of the opinion that the engine is running too rich. I'm not able to bench test the MAP sensor as I don't have a 5 vdc power supply or a handheld vacuum pump. The cost was reasonable enough to just get it and try it. I'm hopeful this could be the issue as I believe it's the sensor directly responsible for trimming the base fuel map based on engine load. In other words, it could be responsible for an over rich A/F mixture if it's out of range. The coil is another concern. If it's weak, I think it's possible it could be causing the same symptoms.

The only other piece of the puzzle I can think to mention is the perforated metal element that was stuffed into the tailpipe, before the bolt on spark arrestor. It was quite badly plugged up with carbon and I opted to remove it completely, which was admittedly a chore. You can-not just pull it out. There is a pin welded to one side of the pipe, through the "media". The only way to get it out was to drill the weld and separate the pin from the pipe. I just used a sheet metal screw to then plug the hole. I have to imagine this "element" was creating, or increasing backpressure on the engine. Maybe that's why it was there, I don't know. This could be the reason, or a contributing factor at least, for why I'm seeing compression readings of 200# instead of the 185# maximum specified in the WSM. That seems "thin" but it's the best explanation I have at this point. I suppose it's possible that the cylinder head had been resurfaced in the past, but I didn't notice any evidence of that. Another concern on this point, if the compression is higher, the MAP may be lower, which would undoubtedly cause the ECU to enrich the A/F mixture. I do have a vac gauge and I will take a some readings for reference. I'm not sure how to overcome that issue, if it turns out to be the problem. Either a suitable, rescaled MAP sensor, or a carburetor retrofit as previously suggested by someone. That will be the extreme and I can only theorize about how to accomplish that. I suppose I would simply remove the injector and figure out the fuel delivery and governor /linkage. I wonder if there are any aftermarket ECU's that could be adapted to allow access to the tune parameters? That would be ideal, but ultra complex I think.

So to summarize, I'm waiting on the replacement MAP sensor and ignition coil and I'll perform a MAP vacuum test to get some baseline figures. I don't recall seeing that value specified in the WSM but I'll take another look for it.

 

[kmelander]

The struggle continues. I replaced the MAP sensor 1st. No difference other than a harder start than usual on the 1st start. Normal after that. Then the coil, which did improve the running characteristics, but not eliminate the misfiring. Under load (driving) full throttle it ran good. Part throttle is still unacceptable. So, I still have not found the gremlin in this engine. I'm wondering about the crank sensor, which is a 5-star PITA to change. I'll need to be certain that's the problem before I undertake that project.

To do that, I'm thinking I'll need to tap into that circuit and log it somehow. Feels like an oscilloscope meter will be required?

 

[kmelander]

Have not figured out the "live" crank sensor test yet. I did do some testing centered around the throttle body. Specifically, I experimented with the fuel evap system and the pcv system. Basically, I removed the vacuum connections for both and plugged them to see if there was any difference in run quality. If there was, it was only slightly better. The engine does not misfire under load anymore. It only misfires under idle, no-load and decel. I think it's actually on the verge of back-firing when decelerating, off load. It just seems to be running slightly too rich and the ecu is commanding this condition for some reason.

This is a function of IAT and MAP, if I'm understanding the fuel injection strategy correctly. I've replace the MAP sensor and measured the resistance of the IAT. I need to revisit the IAT and check it's calibration again. Theoretically, if the IAT was reading low, the ecu would enrichen the AFM. Same for the MAP. I can't rule out the MAP sensor either because this engine has freakishly high (7.5% above factory spec) compression readings. It seems possible that alone could be telling the ecu to enrichen the AFM.

I'm planning to test the calibration of the IAT. If it passes, I may try to fix or offset it's signal to the ecu using precision resistors in an attempt to minimize the enrichment calculation and see what results on run quality.

 

[kmelander]

I checked both IAT and ECT sensor's. Plotted their range in Excel and they are fine. I attempted to splice a vacuum gauge into the MAP circuit to see if I could get any readings but the engine would not start. I guess the volume of the added hose and gauge altered the signal too excessively. The gauge did bounce between positive and negative values during cranking. Is that normal? Positive pulses in the intake? I was expecting chaos but only in the negative region of the gauge. I wonder if I need to take another look at my valve timing and clearances. This will make it 3 times.

 

[kmelander]

Today is 7/14/2024. I started working on this thing mid-April. That's about 3 months I've been chasing whatever gremlin exists in this engine. To recap:
1. Bought it with a blown head gasket for $1k.
2. Changed the head gasket (about $500 with other consumables), was misfiring badly.
3. Started chasing the misfire - spent another $600 on new/used parts to include spark plugs, fuel injector (new), map(used), ECU(used), coil(used). Thank you to Muffs Motorsports on ebay for selling the used parts at a reasonable cost + fast shipping.

The issue has improved with all of this work and replacement parts, but has not been eliminated. I've done a bunch of testing, and need to thank my local dealer for having the kindness to provide the WSM electronic file for this machine.

I've tested compression, fuel pressure, temp sensors and verified cam timing and valve lash adjustments. The only things left, that I can think of, is the crank pulsar and the correct spark plugs. NGK BK4RE plugs were a bit of challenge to acquire. The ones I put in were an Autolite cross reference. I have them now and will install them when I have time and motivation but doubt it will resolve the issue. The crank pulsar is a tough component to test and a big project to replace. It's a massive endeavor, which I lack the interest and motivation currently to pursue. I'd much rather delete the inaccessible ECM and install an aftermarket solution that provides access to the logging and parameters. This will require some fabrication to add a crank position sensor, throttle position sensor and wide band O2 sensor. My ideas involve replacing the throttle body with an Asian donor piece that has a TPS, attaching a crankshaft reluctor to the cam timing belt side of the crank, and welding an O2 bung into the exhaust. ECM would likely be a speeduino type as it's cost with the software, is very reasonable. This is a large project as well and I've already got a very large automotive resto-mod project underway, which takes priority. I did a fair bit or research on this topic and found this YouTube channel/video series was very informative.

Anyways - chalking this purchase up to a loser at this point. Not a complete loss, but for the moment I'm behind on it. I've learned much, and I've shared much. This thread (on this machine and topic) is probably one of the more informative in existence on the internet, which is really weird to me. I can't explain why. I've searched and found practically nothing, anywhere. Thanks to those who have tried to help. I'll try to keep this thread updated as developments occur.

 

[lugbolt]

this is an instance where you REALLY need a way to see what the ECU sees as far as sensor data. I don't recall being able to do that even with kubota dealer's software (Diagmaster).

have you addressed the coolant temp sensor? It also plays a critical role in the EFI's strategy.

 

[kmelander]

this is an instance where you REALLY need a way to see what the ECU sees as far as sensor data. I don't recall being able to do that even with kubota dealer's software (Diagmaster).

have you addressed the coolant temp sensor? It also plays a critical role in the EFI's strategy.

yes I did check the ECT. Post 29.

 

[kmelander]

I decided to dig out my timing light and see what I could see with it. I didn't take the timing cover off or anything, just hooked it up and let is strobe while the engine idled. It did not look proper in my opinion. It had a pattern of slow with miss, followed by fast and then back to slow/miss again. The cycle just repeats. I'm not sure what that tells me. The engine RPM isn't changing as the timing strobe suggests so, maybe a problem with the flywheel? Maybe the key has partially sheared or deformed allowing the flywheel to move on the crank? Maybe the pickup is defective? Maybe the ECU is scrambled? The only thing I can think to do is test the trigger sensor with an oscilloscope and see if it's varying wildly, assuming thats even possible.

 

[kmelander]

Anyone think this little $2 part is my issue? Or maybe it's the coil (#050)?

Either way, it's a lot of work to even look at them.

 

[lugbolt]

flywheel key could be an issue

if it's "that" far out, though, normally you can put a cylinder at TDC and look at the timing mark. If it's off a little the key would be suspect. On those, though-because of how kubota designed it, that's not an easy task to do as I recall.

 

[kmelander]

flywheel key could be an issue

if it's "that" far out, though, normally you can put a cylinder at TDC and look at the timing mark. If it's off a little the key would be suspect. On those, though-because of how kubota designed it, that's not an easy task to do as I recall.

I actually considered using a small hole saw to bore a hole through the transaxle housing. I just want to know if the flywheel is wandering on the crankshaft. I'm fully prepared to install the MicroSquirt ECM, which involves a new crank trigger anyways. If the flywheel is destroying the crank, I've got to deal with it either way.

 

[lugbolt]

microsquirt...oh fun.

had one on my old SVO. Pretty easy, for the most part, and in the case of the svo, opens up a TON of options that the stock PE ECU wouldn't support, such as bigger injectors, coil-near-plug (and individual cylinder timing) and removal of the vane air meter which is a restriction. That in itself is worth 30-40hp. Took a ~280hp 2.3L to about 400hp along with other mods of course. Pretty fun little turd to play with and put the beating on some unsuspecting V8's at the drag strip

you'll want to read into it though because some things aren't included with microsquirt, and I don't know of the rtv might need it or not. diyefi.com i think is what i used.

getting the fueling and timing correct is the hardest part and that's no big deal-once you figure it out. And RTV doesn't have an oxygen sensor, so you can add one if you need to-or you can run open loop all the time. I like having a feedback system though, IF that's possible on the RTV, being that it has a kitty cat in the exhaust manifold.

on the rtv you'll set it up as waste spark, it'll fire both plugs at the same time. easy.

Lots of reading to do on the subject, and I am not sure anyone's ever done it so you may be on your own with it as far as getting much help, in terms of figuring out the crank sensor types, etc

but it's not rocket surgery.

 

[kmelander]

microsquirt...oh fun.

had one on my old SVO. Pretty easy, for the most part, and in the case of the svo, opens up a TON of options that the stock PE ECU wouldn't support, such as bigger injectors, coil-near-plug (and individual cylinder timing) and removal of the vane air meter which is a restriction. That in itself is worth 30-40hp. Took a ~280hp 2.3L to about 400hp along with other mods of course. Pretty fun little turd to play with and put the beating on some unsuspecting V8's at the drag strip

you'll want to read into it though because some things aren't included with microsquirt, and I don't know of the rtv might need it or not. diyefi.com i think is what i used.

getting the fueling and timing correct is the hardest part and that's no big deal-once you figure it out. And RTV doesn't have an oxygen sensor, so you can add one if you need to-or you can run open loop all the time. I like having a feedback system though, IF that's possible on the RTV, being that it has a kitty cat in the exhaust manifold.

on the rtv you'll set it up as waste spark, it'll fire both plugs at the same time. easy.

Lots of reading to do on the subject, and I am not sure anyone's ever done it so you may be on your own with it as far as getting much help, in terms of figuring out the crank sensor types, etc

but it's not rocket surgery.

"kitty cat in the exhaust" = not any more - assuming you are referring to the metal packing in the tube that was pinned in just before the spark arrestor. That went bye-bye already.

I've been reading about and researching the MS a good bit already. This is "escalation of commitment" at it's finest.

 

[Nicksacco]

I'm not an expert on this machine, but a couple of things kept bothering me as I read through this (multiple times!)

The Cylinder head:
Did you ever get an accurate gauge on bolt torque? 14 ft/lbs doesn't sound unreasonable for a little engine.

You mention the compression is much higher than spec. Do you think you're getting detonation?
Since you removed the head, you know there's no carbon buildup. but what about valve train? Could the timing of valve train be off causing the high compression? (assuming the valves are adjusted correctly).

Ignition:
My first thought was coil and/or wires also but it seems you've replaced spark plugs/coil etc. to no avail - is that right?

Does that engine use "wasted spark"? I couldn't figure that out from this thread? Could the wasted spark be causing detonation or similar if the valve train timing were off?

Is the spark system getting enough power? Have you tried a new(er) battery? Are all grounds clean and good? Battery terminals clean?

Fuel:
I think you say the engine is always running rich which is showing as blackness on the spark plugs.
Since this thing has a computer controlling it, what would make the computer think the engine was too lean? Temp sensor would do it - possibly stuck thermostat?

Is the gas in the tank old/have water in it?

I hope this helps and I hope you get this figured out.

 

[kmelander]

I'm not an expert on this machine, but a couple of things kept bothering me as I read through this (multiple times!)

The Cylinder head:
Did you ever get an accurate gauge on bolt torque? 14 ft/lbs doesn't sound unreasonable for a little engine.

You mention the compression is much higher than spec. Do you think you're getting detonation?
Since you removed the head, you know there's no carbon buildup. but what about valve train? Could the timing of valve train be off causing the high compression? (assuming the valves are adjusted correctly).

Ignition:
My first thought was coil and/or wires also but it seems you've replaced spark plugs/coil etc. to no avail - is that right?

Does that engine use "wasted spark"? I couldn't figure that out from this thread? Could the wasted spark be causing detonation or similar if the valve train timing were off?

Is the spark system getting enough power? Have you tried a new(er) battery? Are all grounds clean and good? Battery terminals clean?

Fuel:
I think you say the engine is always running rich which is showing as blackness on the spark plugs.
Since this thing has a computer controlling it, what would make the computer think the engine was too lean? Temp sensor would do it - possibly stuck thermostat?

Is the gas in the tank old/have water in it?

I hope this helps and I hope you get this figured out.

Thanks for taking an interest. I'll try to answer all the great questions below, in the order you asked.

1. I did get the head torqued to factory specs finally. I had to replace all of the head bolts with new and had to buy another new head gasket to get this correctly repaired.

2. Using a standard compression tester, I'm getting 200# from each cylinder. The shop manual specs 185#. I don't think I'm getting any detonation though. Just an annoying misfire.

3. I'm fairly certain the valve timing is ok. I've adjusted the lash twice now and verified the timing marks on the crank and cam. I do plan to check the lash adjustment one more time though because I've learned of a different method which seems (I think) to be better. None the less, it's making 200# compression so I'm thinking it's probably not totally out in left field. Unless the lash setting is somehow causing the high compression as you mentioned, which I'm struggling to understand how that could happen, but I'll admit it may be possible.

4. I have replaced the spark plugs, injector, ecu, coil & wires and MAP sensor. I do think it has to be a wasted spark ignition because it's a single coil with 2 plug wires. I guess it would cause poor running conditions if the valve timing were off. It doesn't run horrible, just doesn't run great. Also, it has improved vastly with the replacement of the ecu and MAP sensor. The ecu, MAP and coil replacements were used from ebay.

5. I'm getting 12+ VDC from the battery standing and 13+ VDC running. I pulled and cleaned all of the ground wires/lugs and battery terminals. Great question here. I admit though, I have not tried a new battery, only because I can't identify a problem with the current one.

6. I did verify the fuel pump pressure is above 36# per the shop manual. I deleted the charcoal canister and removed the tank, removed the pump and cleaned both refilling with fresh ethanol free gasoline.

I also hooked up my timing light to see what the spark timing looked like, which looked very erratic. It wasn't what I was expecting. Basically, I'm down to thinking that there may be a problem with the crank trigger or flywheel. Problem is, I believe the entire rear carriage has to be removed from the buggy and then the engine has to be separated from the transaxle just to gain access to those components. I'm pretty much waiting for the fall to attempt that project. I really like to avoid it all together by just abandoning the factory engine management system. The one caveat with that plan is if there is a problem with the flywheel key, I still need to address it regardless so...I don't think I can avoid tearing it all apart. I think what will likely happen is I'll break it down to get all the modifications necessary for the aftermarket ECM done, and I'll check the flywheel while I'm in there. Its really tedious and annoying trying to troubleshoot this problem with no access to the ECU so given the choice, Ill abandon the OEM computer without much concern at this point. It's been a really odd experience trying to get information about repairing this apparent unicorn side by side. Doing the usual YT and internet searches come back with practically nothing. That never happens. But the older RTV-500 seems to be the exception. I still think it's a cool little machine but now it's simply a battle of will. Challenge accepted!

Thanks again for taking an interest. Let me know if you can think of anything I'm overlooking. Best wishes.