Cold starting problems

Guest Discussion
1

I think its time to raise the yellow caution flag here before someone’s airplane burns to a cinder.

Continental engines are normally VERY easy to start cold. In fact, I had one with a bad starter on a ferry trip in Africa where it had to be hand propped for three straight mornings. This was a three blade prop in a Cessna 206 and it fired up each time in one blade. As to who would prop an IO-520 with three blades out there, it wasn’t me!!

It is time to come out and just say that something is wrong in the starting process of this engine and it ought to be addressed up front by Cirrus/Continental.

I base this observation on several facts:

  1. The problem Rob Leach and I had starting N142CD in Ogden, UT. We created a huge lake of fuel under the cowling and were told by the factory to “prime some more” as we ran the battery flat. This was a definite fire hazard. We even tried a flooded start procedure with the same result - no cylinders fired. There just did not appear to be any fuel in the cylinders despite the flood on the ramp.

  2. After the engine change, this same plane has had the same problem in Australia per Clyde’s recent post.

  3. Walt has had the same problem in California.

  4. Han has experienced it in Europe.

There is no way that you should have to run down a battery or create a fuel puddle below an airplane engine to start it. Something is wrong.

This engine has a top induction system unlike most of the other larger TCM engines. Whether that is responsible for its strange start characteristics, I don’t know. What appears to be happening is that when priming occurs, fuel is flowing but evidently not to an effective place for starting. Additionally, Walt reports something very strange – when he primes, the fuel flow reaches a peak and then goes to zero – this with the pump on, the mixture rich and the throttle full forward. I have never seen this occur in a Continental engine. With the controls in this position, fuel flow should be continuous. Evidently something is shutting off fuel flow to the cylinders but allowing it to flow elsewhere in the engine in order to reach the overflow drain.

Since I have neither a Cirrus at hand nor a maintenance manual, there is not much I can do to investigate this matter further. However, absent any insight from the manufacturer (and there ought to be some very soon, I hope), I would suggest the following procedure as an experiment:

For your next cold start, begin cranking the engine at the same time that you see the first indication of fuel flow from the primer. By doing this you will be sucking the prime vapor into the cylinders immediately, giving the engine a chance to fire while fuel is definitely there. It seems that by using the factory method, fuel vapor is draining away before combustion can occur. In short, it is like trying to start the engine on air alone!

I would appreciate any feedback on how this works for you. The current situation is abnormal and unacceptable. Unless it improves, someone’s airplane is going to be toast.

2

It is time to come out and just say that something is wrong in the starting process of this engine and it ought to be addressed up front by Cirrus/Continental.
After talking to Cirrus, and getting feedback from Walt and Han, I don’t believe there is a real problem, but there is a difference between this engine and e.g. a Cessna 206. The difference is that because the plane is certified under FAR23, it is a certification requirement that operation of the priming pump during flight should not stop the engine. While I’ve never tried it, I’m told that operating the priming pump on older fuel-injected aircraft while the engine is running will make it stop from an oversupply of fuel.
Consequently, the flow rate of the primer has been limited on the SR20 to meet this requirement. This is why the fuel flow spikes then drops to almost nothing (the spike is presumably the pump pressurizing).
There does seem to be a need to revamp the POH and other documentation to revise the recommended starting procedure. Using the “normal start” when the engine is not already warm does not seem to work. I’ve been told by Cirrus to forget about the “normal start” and treat every start as either a hot start or a cold start.
The key to the cold start procedure seems to be to barely crack the throttle, whereas the manual suggests opening it 1/2 inch or so. I’ve had confirmation from three separate sources that using this technique eliminates the problem completely. I guess basically the effect is to limit the amount of air entering the manifold, so as to produce a rich mixture with the available prime fuel. Opening the butterfly valve even a little, at cranking speeds, is going to admit a lot of air.
As to the fuel puddle that can develop under the aircraft, I agree that this is a fire hazard, but using the correct technique it doesn’t seem to happen. Exactly where it’s coming from is not clear, but the 10-12 seconds of prime that is recommended for a cold start doesn’t seem to cause any overflow.
So, in summary, this engine is different, but not intractable given the right handling, but the documentation needs to be improved to reflect this.

3

I second the Clyde comments. There is no real problem, only if you follow the POH. Recap what works for me (and it fires up in at first or second cranking):
Cold start: 12 second prime (throttle forward), retard to a crack open, crank and “play” with the throttle within a crack and 2xcrack limits. Outside (too much or too little) causes the engine to die (from overdose or starvation). Keep thumb close to boostpump, as it may help smoothing it out.
Warm start: Same procedure, but only 3-5 second prime.
I have never caused or seen a puddle. If you would have to keep priming, I’d say you’re not reaching the spot where the fuel should go, simply overflowing it… This would be highly suspect AND dangerous!!
I’ll keep an eye on this (as everyone else I presume).
With regards from boring Paris.
HK

It is time to come out and just say that something is wrong in the starting process of this engine and it ought to be addressed up front by Cirrus/Continental.

After talking to Cirrus, and getting feedback from Walt and Han, I don’t believe there is a real problem, but there is a difference between this engine and e.g. a Cessna 206. The difference is that because the plane is certified under FAR23, it is a certification requirement that operation of the priming pump during flight should not stop the engine. While I’ve never tried it, I’m told that operating the priming pump on older fuel-injected aircraft while the engine is running will make it stop from an oversupply of fuel.

Consequently, the flow rate of the primer has been limited on the SR20 to meet this requirement. This is why the fuel flow spikes then drops to almost nothing (the spike is presumably the pump pressurizing).

There does seem to be a need to revamp the POH and other documentation to revise the recommended starting procedure. Using the “normal start” when the engine is not already warm does not seem to work. I’ve been told by Cirrus to forget about the “normal start” and treat every start as either a hot start or a cold start.

The key to the cold start procedure seems to be to barely crack the throttle, whereas the manual suggests opening it 1/2 inch or so. I’ve had confirmation from three separate sources that using this technique eliminates the problem completely. I guess basically the effect is to limit the amount of air entering the manifold, so as to produce a rich mixture with the available prime fuel. Opening the butterfly valve even a little, at cranking speeds, is going to admit a lot of air.

As to the fuel puddle that can develop under the aircraft, I agree that this is a fire hazard, but using the correct technique it doesn’t seem to happen. Exactly where it’s coming from is not clear, but the 10-12 seconds of prime that is recommended for a cold start doesn’t seem to cause any overflow.

So, in summary, this engine is different, but not intractable given the right handling, but the documentation needs to be improved to reflect this.

4

I second the Clyde comments. There is no real problem, only if you follow the POH.

Han, you’re a natural comedian. I just burst out laughing when I read this. The only real problem occurs when you follow the factory procedure…

I thought the opposite was supposed to be the case.

Enjoy your stay in Paris. :slight_smile:

5

In that same vein (it’s late TGIF here) there is one other possibility why Clyde’s cold starts are different: The European air is somehow different than Down Under, or gravity… Do I smell Chanel 5 in the air here??

HK

I second the Clyde comments. There is no real problem, only if you follow the POH.

Han, you’re a natural comedian. I just burst out laughing when I read this. The only real problem occurs when you follow the factory procedure…

I thought the opposite was supposed to be the case.

Enjoy your stay in Paris. :slight_smile:

6

In that same vein (it’s late TGIF here) there is one other possibility why Clyde’s cold starts are different: The European air is somehow different than Down Under, or gravity… Do I smell Chanel 5 in the air here??

HK

Maybe it’s the beaujolais, a smidge of alcohol vapor should help the sr20 start right up.