11 replies to this topic

[Soupbones]

So I'm a little naive on this subject so feel free to teach me............

I've read post that mention fuel/air ratio and am interested to find a way to monitor as I ride.

What is normal reading for a 1000 SDI 2006 Mid Altitude MAP at idle, partial, and full throttle?

 

This is my second rebuild with the mag side piston failure due to detonation and skirt scored so this time around I will ride with a fuel pressure gauge and a Erave micro switch/LED. Sent my injectors to Injector RX..........FYI great service and cost.

 

[Daag44]

I'm sure you'll get some help on it. You can also get all the numbers from your own sled.
 

[nm9stheham]

Not sure this will help the OP but here goes.....I read over this particularly good thread: https://www.dootalk....=air fuel ratio

But I did not see one thing discussed that has an impact on the AFR readings from an O2 sensor in the exhaust with a 2 strokes in particular.

 

Anytime that there is unburned fuel passing from the intake to the exhaust, an O2 sensor is going to read lean. This happens for 4 strokes when warming up, and with huge cams that give a lumpy idle. It happens for 2 strokes when warming up AND when you are hard on the throttle and the scavenging is carrying some of the intake fuel-air mix out the exhaust.

 

The reason is this: An AFR gauge that uses an O2 sensor only measures the % of free oxygen in the exhaust, and interprets that as AFR in the intake. When full combustion of the ingested fuel-air mix is going on, then the free oxygen content in the exhaust is an accurate reflection of the fuel-air ratio in the intake. But when some of the intake fuel is not burned, then the oxygen that goes with that unburned fuel is also not burned, and that unburned oxygen shows up in the exhaust as free oxygen. This excess free oxygen in the exhaust is interpreted by the AFR gauge as a lean condition in the intake fuel-air-mixture.

 

So I would not expect an O2 sensor-based AFR gauge to give accurate readings under heavy loads, big throttle openings, and higher RPM's on a 2 stroke. (And actually, this has been known for a long time...) I would expect the readings to be lean (high). It will read accurately at low to mild throttle and RPM, and idle after the engine is warmed up. The seemingly high AF reading posted by spookum in post #25 of the above linked thread is very probably explained by this matter.

 

So the only use for an O2-sensor-based AFR gauge for the OP under hard throttle conditions is to see if there are sudden and very large spikes upwards in the AFR readings. The typical readings at hard throttle won't be accurate. (EGT sensors for 2 strokes are used for a reason.....)

 

And detonation can be caused by other things than air-fuel ratio. Excessively high head temps is one good cause.

[Daag44]

Do you remember what rpm and throttle position the mag side burnt down? The throttle position doesn't need to be precise, 1/4, 1/2, 3/4 o full throttle.
 
Since it was the mag side piston that failed both times, this also means the pto side survived so it is worth comparing both sides carefully. Some things to check are the combustion chamber volume, the injector flow, crank phasing, and the EGT temps. I wouldn't be too concerned with minute differences, but rather look for something that stands out like the mag side running 100F hotter than the pto side.

I would also inspect the muffler carefully by shaking it to look for any loose baffles that could increase the backpressure.

[Daag44]

Nm9stheham, I think that you meant the A/F is near accurate at full throttle, and misleading at low to part throttle due to having more air than fuel short circuiting out the exhaust. When running a Wideband on a two stroke, the reference I use is what works. I'm not too concerned with the relation to Stoic, because all two strokes set by the manufacturers run on the Rich side of stoic due to emissions. The only Rotax that I know runs Lean of Stoic is the 912i which is a 4S that runs Lean Of Peak for Fuel Economy. It can and has been done on two strokes even with carbs, but I haven't seen anyone take a serious interest in this type of tuning. You can imagine the trouble with carbs to transition from Lean to Rich. It is better suited for injection like our sleds.

[nm9stheham]

Nm9stheham, I think that you meant the A/F is near accurate at full throttle, and misleading at low to part throttle due to having more air than fuel short circuiting out the exhaust. When running a Wideband on a two stroke, the reference I use is what works. I'm not too concerned with the relation to Stoic, because all two strokes set by the manufacturers run on the Rich side of stoic due to emissions. The only Rotax that I know runs Lean of Stoic is the 912i which is a 4S that runs Lean Of Peak for Fuel Economy. It can and has been done on two strokes even with carbs, but I haven't seen anyone take a serious interest in this type of tuning. You can imagine the trouble with carbs to transition from Lean to Rich. It is better suited for injection like our sleds.

Thanks daag. Just to make it 100% clear, it is not at all matter of more air than fuel short-circuiting out the exhaust .... it is the unburned fuel-air mixture going out the exhaust in the same mixture proportion as coming into the cylinder. The O2 sensor cannot sense unburned fuel; it only sees the short-circuited oxygen in addition to the rest of the normal free oxygen with the exhaust gases.

 

You are very probably spot on for the conditions under which this occurs.... my 2 stroke knowledge is pretty weak compared to 4 stroke. So to say it better, under whatever conditions the scavenging carries some of the intake fuel-air mixture through the chamber unburned, and on out the exhaust, is where this lean reading will occur with an O2-sensor-based AFR gauge. I thought that occurred at WOT.....but I need to do some homework and get better educated there.

 

For most 4 stroke operation, once up on the torque curve at WOT, the fuel-air is being well ingested and burned in the cylinder.... that is why you are up on the torque curve LOL. So that makes sense for what you say.... if you are at WOT and well up on the 2 stroke torque curve, then it is for the same reason: the fuel-air mix is being well ingested and burned.

 

I saw you posted some AFR numbers from an engine on a dyno and figured up the AFR from measured air and fuel flows. If it could ever be arranged, it would be very good info to take an engine characterized like that and put a O2 sensor-based AFR on the engine while on the stand, and compared the results. And then take the engine in a sled and see what the O2-sensor-based AFR gauge says in real operation.

 

I would not be surprised if that comparison info was somewhere....

[Daag44]

Injecting fuel from the top of the transfer ports - or even above them - provides a way to let intake air to enter the cylinders before having to pulse the injectors. The faster the engine turns, the less time there is to stay ahead of the short circuit. The E-TEC has the advantage of even more time to stay ahead of the short circuit with both stratified and homogeneous charge, but it too runs out of time as the engine turns faster and requires more fuel.

The oxygen measured by the Wideband will be the sum of the air that short circuits out the exhaust prior to combustion and the surplus from a Rich combustion, minus the oxygen from any further catalyst in the exhaust prior to the placement of the Wideband. BRP's savvy engineers calculate the actual a/f ratio in the combustion from the level of emissions.

[maxmachz]

Back to the original question...........Unless you are riding around constantly looking at this (or any) gauge, most likely the damage will happen before you know it.

 

If you can figure out a way to have a green light, then yellow, then red that you can see peripherally as you fly down the trails you may have a slim chance of mitigating damage. Most likely your issue that caused the failures happened pretty quickly, if you saw anything on any gauge it would just be telling you it's already damaged.

 

They are great for dyno tuning because you can stare at it and other monitors as you run an engine through the rpm range, and make incrementally small throttle adjustments watching closely for anomalies; but out on the machine / on the trail........not so much. 

 

If you want one for tuning, and have a place to run the machine and not have to look where you are going, that's another story and it surely may be helpful. 

[nm9stheham]

It's really hard to say if you would catch it or not. Detonation is not always instant damage.... at least in the auto engine world. You might catch some erratic trends. But the sensor better had be in the failing cylinder's exhaust alone.

 

Dumb question: Do these engines have knock sensors? A light tied to one, with a 'hold' function (keeps the light on after 1 or a few knocks), would be the most instantaneous sensor.

[Daag44]

It has a Knock Sensor that gets used from 5,000 rpm and up. The later calibration has increase sensitivity. I know it works on the E-TEC, but for the SDI it remains a mystery. The tuning aspect is relatively easy. Within 5 min you can get a pretty good idea. The two scientific measurements that I use are In the Ball Park and Out in the Left Field. From Spookum's readings the one at low load appears too lean, so I would use the EGT readings to find the cause. Once the combustion between cylinders are relatively close and the A/F is in the ball park, then I go ahead and do plug readings in a few different load range to confirm all is good.
 
"The ECU doesn't look at the knock sensor until 5000 RPM on the 1000 SDI 2 stroke."
https://www.dootalk....e/#entry3694299

[Daag44]

It's really hard to say if you would catch it or not. Detonation is not always instant damage.... at least in the auto engine world. You might catch some erratic trends. But the sensor better had be in the failing cylinder's exhaust alone.
 
Dumb question: Do these engines have knock sensors? A light tied to one, with a 'hold' function (keeps the light on after 1 or a few knocks), would be the most instantaneous sensor.

 

Klemm Vintage Racing Engines has a good web page on detonation management for two strokes. Making sure the engine is running right will go a long way to prevent detonation. With two consecutive mag side burndown I would measure the a/f and compare the EGTs between the mag and pto.

 

Detonation Management – The Key to 2stroke Power and Reliability

a technical paper by Harry and Gerhard Klemm  of Klemm Vintage Racing Engines

https://www.klemmvintage.com/deto.htm

[800PTEK]

Have you inspected the y-pipe? They have an inner lining that was known to collapse and cause issues. May be worth a look. Its just another thing that would consistently be linked to that one side of the engine.Do you also do a crankcase leak check after a rebuild? A bad crank seal or a delaminating reed boot would also fit. All of these I would believe could give you readings that are inaccurate.