Observations on Classic FWE - Suggestions?

[larry cottrill]

I mentioned somewhere earlier that I had finally managed to ding up the original FWE prototype intake I used [1/2-inch conduit] to the point where I could no longer get sustaining operation. I had also tried shortening the tailpipe, since Eric had good luck with this.

I decided a while back that this should be the engine that I use to test the potential of the oxygen sensor as a research tool, so what I did was add an M18x1.5 flange nut near the tail end of the cone, and at the same time cut out and replace the original intake with a properly sized and flared one [3/4-inch conduit, like my original design and as Steve used on his]. All this was put in place Saturday. To my astonishment, however, I was unable to get sustaining operation!

I experimented with it extensively yesterday, after adding a typical starting air / fuel tube assembly, which at least eliminated the need for three hands while starting. To my amazement, the engine will only sustain when I extend it to the original design length - two inches [50 mm] short will NOT do! Note that I was very careful to build and locate the intake pipe according to my original design, including a fair amount of pinch at the inside end.

Apparently, the length ratios are very critical on this engine as it now exists. I will need to permanently add the tailpipe length back on for it to start and run reliably. I really wanted to get this engine, with its "amazing mystery probe" O2 sensor, sent off to the First Great International Pulse Combustion Cookoff and Wine Festival, but I'm having a hard time coming up with a sturdy carton of sufficient length for the full size engine. I'll do my best ...

Now, here's the deal: I wouldn't mind putting something a bit different on the tail end - but, I don't have a lot of time to play around. I could make a small cone and weld it on, or a small cone with a flared inlet and a small gap a la SNECMA. Or, a larger tube with a flared inlet and annular gap a la Rossco. Or, cut it back more and put on a long cone! Or something else. It could be an "augmentor" style device, as long as it's kept closely coupled to restore the effective length. I just have to do something to add the acoustic length back in, and it has to work the first time. Performance impact is of no real consequence, and can be measured later. Note that for me, it would be no big deal to add something with a small gap and simply weld it in solid if that didn't work! I can test fire almost immediately after modifying, so the development cycle should be short if the mod is simple.

So, gentlemen, tell me quickly: what would you like to see tried?

L Cottrill

[Eric]

Larry what the heck did you do to the engine? I have had them run even with 14" tailpipe lengths! Whats your intake length?

Do you have the fuel probe in the right spot? That alone can make a world of difference.

I need to send you one of my SUPER FWE's and then you can mod from their, although they are pretty damn dialed in as is and any gain would be minimal. With augmenters it could replace the dynajet.

I am going to go do some CL flying this weekend, and also run some engines on a teather car so I will let you know how it works out. Ill have lots of pics and videos taken.

ACK! I just noticed you said you pinched the intake! ..... can you un-pinch the intake?

Eric

[larry cottrill]

Larry what the heck did you do to the engine? I have had them run even with 14" tailpipe lengths! Whats your intake length?

Do you have the fuel probe in the right spot? That alone can make a world of difference.

I need to send you one of my SUPER FWE's and then you can mod from their, although they are pretty damn dialed in as is and any gain would be minimal. With augmenters it could replace the dynajet.

I am going to go do some CL flying this weekend, and also run some engines on a teather car so I will let you know how it works out. Ill have lots of pics and videos taken.

ACK! I just noticed you said you pinched the intake! ..... can you un-pinch the intake?

Eric - don't panic!!!

All I pinched on the intake was the forward end where it spouts into the engine - about like Steve did, or maybe just a bit more. So no, it can't be undone, but I don't think it's the issue. It is a fully round section until below where it makes the bend down into the chamber. The engine runs fine once you get the pipe length tuned. The intake total length is just a hair short, but within a small fraction of an inch of the way it was designed. Alignment of the intake pipe inside the chamber seems to be pretty good. The heat pattern appears to me to be just as in Steve's prototype photos.

The fuel pipe location is dead center, spouting just inside the throat where the flare narrows to the normal ID of the tube. I could rig a new fuel tube that would go in deeper, of course, but that has worked for me before. The intake tube is around .85 inch ID, the fuel pipe is .125-inch OD copper, as usual. There is very little intake flame ejection, even when running pretty hard.

As I said, it runs fine once you get it tuned. The only thing that bothers me is how critical the length seems to be, which is obviously at variance with your findings.

What are you going to fly control-line - the Super FWE? Man, would I like to see that !!! What sort of machine are you strapping it to - something I've heard of, or a custom plan form of your own?

L Cottrill

[Eric]

Well all the engines that I pinched the intake had no where near the operating range or performance as a straight in engine.

The pipes that let the air flow right in nice and smooth you can really abuse and do all kinds of drastic modifications and the engine will just keep running. The one engine I started off with 19" tailpipe lenght and locked it up in the vice right next to the chop saw, ran it, chopped half an inch off, deburned it quick with a knife and started it right back up again. Below 14" things start to get unreliable, 12" seems to quit altogether but probably because the intake is way too short. I would be willing to bet the total length could be reduced to about 14" total if multiple smaller intakes were used, although it wouldnt be optimum.

16.5 - 17.5 seems to be the best range.

Im not sure what I am going to fly and what I am doing to put on the car. for the stuff that is more experimental prototype type stuff I will probably use the car to reduce the chance of a catostrophic crash. I have a few dynajets that I want to try out and a linear valveless pulse/ram jet, and a ducted fan motor jet engine.

I made up a little dynajet speedster plane, and maybe I will finsih up my old ironsides plane before the weekend, but most likely not. The car is totally custom, kind of similar to what you would launch from but much faster. There will probably be a few old ironsides type planes flying and a few other custom ones.

The super FWE just got even more super. I employed some of the supercharging techniques I have been working on and put a massive augmenter on the intake (a left over FWE cone section flared at the intake) and just having the intake augmented total thrust is about 3 lbs. With the exhaust augmenter it should put it well over 4, hopefully closer to 4.5.

Eric

[larry cottrill]

Tonight I found a piece of junk motorcycle muffler and blacksmithed a crude augmentor which I thought might act as the needed extension if closely coupled to the tail end of the pipe. Although it was closely coupled in length, there was a lot of slop in the difference in size between the tailpipe and the add-on. The thing totally failed to act as an extension, even though the acoustic length was theoretically satisfied.

My conclusion is that if the tailpipe abruptly opens into a larger pipe, the engine still senses the abrupt end as the pressure node. Of course, I have no idea what would have happened if there had been a bulkhead instead of an open front end to the add-on - the story might be completely different. Anyway, if the larger pipe is sufficiently "open" front and rear, it fails to behave as a proper extension. This will perhaps seem obvious to many, but it was interesting to try. At least, I didn't make the mistake of welding it on!

This leads me to believe that Bruno's criticism of my "Cottrill Augmentor" posted on the old forums is probably wrong! Bruno, you'll remember saying that the augmentor would need to be spaced pretty far to the rear to prevent unwanted acoustic coupling. I now believe this is not true, IF the nose of the augmentor is sufficiently large to offer plenty of pressure relief around the tailpipe end. Of course, your criticism that it might inhibit proper tailpipe breathing could still be correct - but now, I have the exact part needed to test it on some working engines! The one I made is slightly larger than shown in the drawing.

See below to refresh your memory ...

L Cottrill

[larry cottrill]

All right - nobody's suggesting anything, I'm running out of time, and I need something simple that's worth trying and that has a good chance of working, AND that I can throw together in a hurry. Tonight's my last shot if I want to get this out to Mike in time for it to show up for the Meet.

So, here it is:

I'll cut a short stub off the flared end of the tailpipe, and then shorten it a bit more. I'll weld the flared stub onto my remaining FWE chamber cone from Steve, then turn this around to form a SNECMA-style "skirt" augmentor cone, and attach it to the shortened tailpipe with a 1/8-inch gap. I'll secure it with three or four little welds, then fire this up to see if it will run. If it does, that's how I'll ship it. If it doesn't sustain, I'll just seal up the gap with weld, verify that it will run that way, and then ship it! I'll be able to salvage the cone for a second FWE later and patch straight tailpipe onto this one to bring it back to stock dimensions. The best of all possible worlds ... right? See the small scale drawing below.

Any comments? Today is your last chance to come up with something more interesting ...

L Cottrill

[larry cottrill]

Well, I finished it, just like the drawing except for using four weld bridges to hold the two parts together instead of three. Naturally, by the time I got done with it, it was well after dark and too late to test it, but I'll ship it out today anyway. Maybe Rossco can get it to go - he seems to be good with engines that have odd shapes, funny gaps, etc.

Total weight came to a mere 35.5 ounces (cough ... choke ...). I had no way of getting perfect alignment between the two sections, but it really isn't off by much, so it should go. The overall inside length appears to be just a hair under 26 inches, with the gap very close to the 1/8 inch shown on the drawing. The gap dimension was just picked arbitrarily out of the air, of course. It took over an hour to make the four little weld bridges, sculpting them very gradually with a very small torch flame, working from the straight forward section over onto and slightly around the flared entrance.

If I think there's time, I'll get a couple of pictures before I package it up for mailing.

L Cottrill

[larry cottrill]

Just to show the relationship between the original FWE design and the one modified with the SNECMA-style tail skirt.

L Cottrill

[larry cottrill]

Just in case you thought the mounting for the O2 sensor I showed on my drawing was an exaggeration ...

L Cottrill

[Mark]

A backgrounder. Note the decimal in the second link compared to the first link. Guess they need a decimal sensor.
Mark
http://www.volksparts.com/o2sensors.htm
How about some wave forms?
http://hostingprod.com/@aa1car.com/library/o2sensor.htm

[mk]

No, we didn't do any more work one the engine yet.

Seems like Mike is going to send it back as it is.

[larry cottrill]

It had three operating modes, which was unusual. The first I call gruntle, because that's a somewhat onomatopoeic description of it. That's when there is internal combustion, but not at resonance, probably the flame sitting at the inner end of the intake. Second mode was in resonance, but ejecting blue flame out the slot. Third was full blast, and not ejecting flame at the slot. It heated up very quickly in that mode.

Two things need to be done to make it run at full bore without air. First, the intake needs to be unsquashed on the inside. The inner squash effectively reduces the intake flow, but maintains the intake mass, so the engine can't breath properly.

My first reaction to this is, c'mon, guys, it isn't like I squashed it flat - there's more area at the spouting end of the intake now than the whole cross-sectional area of the 1/2-inch EMT I used originally. This engine ran fine with a straight extension restoring its original tailpipe length.

HOWEVER, it is true that the cold piston mass in the tailpipe is now significantly larger than when it was a straight pipe, of course. So, what you're saying here may have merit. What it might amount to is a bad impedance mis-match between intake and tailpipe. Unfortunately, and as usual, this would be a real bear to change now.

Second, the tail cone needs to be cut off and overlapped with the tailpipe. It does acoustically couple as is, but the small starting diameter of the frustum encourages the flame to escape.

Ben, if that was a perfect explanation, you could never get to mode three in your description above - which obviously is the only condition at which the engine was close to running right. To me, that's the most significant observation - you HAVE seen operation where the machine behaves [acoustically] as if the gap were not present! It seems more likely to me that leakage out of the gap indicates extended combustion time with excess static pressure, exactly like when you see an engine with a big, blooming tail flame instead of a nice straight flame. I'd be willing to bet that at the moment you observed that, there was also a lot of flame ejection out the intake ...

One of the welds holding the cone did not get adequate penetration and had already separated from the cone when the box was unpacked. I don't know if they intend to work on the engine yet this week, but I did not want to cut up one of Larry's engines.

I hate it when that happens ... now I'm really ashamed of myself. That probably comes from working in a hurry, late at night, tired, etc. It shouldn't be surprising that the quality differs among the four welds - they were each built individually with a fresh start and adjustment of the torch. It's difficult work because you have to use a very small flame to keep from destroying the edges nearby, especially the lip of the flare. But, I am disappointed that one of them was really that bad.

Well, my thanks to you, Ben, and to everyone else who worked at it. In the valveless pulsejet game, you just can't win them all, I guess.

L Cottrill

[Graham C. Williams]

Dear Larry.
The cone when acoustically coupled to the main exhaust pipe seems to have the same effect as adding a straight pipe somewhat longer than the cone. The apparent increase in acoustic length is exaggerated as you increase the 1/2 angle of the cone. I have recent experience of this effect with the Type07 motor. The all-important location of the pinch points moves (acoustically) as you change the conical ½ angle. Makes it a swine to tune until you realise what’s going on.

Regards
Graham.