breathe in, breathe out

In September, I wrote a very long text-heavy post about the process of internal combustion engines and what things can be adjusted.

I'm getting back to that, so I've attempted to produce a diagram that partially illustrates the process here:


Each thing in parentheses can be adjusted; some things only as part of the design of the engine, others are adjusted when the engine is running. If any of the items fails to work, the engine will not run.

Adjusting the performance of an engine (of given displacement) involves changing one or more of the items in the diagram. The engineering challenge is that such a change also requires you to change other aspects of the engine to compensate.

For instance, the ignition timing and mixture are intimately involved with each other; they need to change in relation to each other for the engine to keep running. If you stomp on the gas and the engine hesitates, it means that when the throttle opens quickly (increase in the effective volumetric efficiency), the mixture and ignition timing aren't changing properly to compensate.

I put in the mechanical removal of heat because that's a critical (and limiting) parameter of the Volkswagen air-cooled engine (like in my beetle). It's essentially a function of the fin area of the cylinder heads and the airflow around the engine. There are lots of ways to increase the power being developed by a VW engine. You can, for instance, make the intake and exhaust system much less restrictive to make the engine breathe better. However, that only increases the load on the cooling system and so makes it easier to overload and overheat the engine.

My next task with my engine is to decrease the compression ratio of the cylinders down to what it should be. I think it's been increased by accident, by people adding used heads without adjusting things properly. That will decrease the burden on the engine's cooling system and allow me to adjust the ignition and carburetion systems correctly, and hopefully stop the exhaust valve from stretching.

holy hyperbaric chamber Batman!

[WARNING: Long explanatory post. For those of you who already know about compression ratios: I think the reason my #1 valve was having problems was the CR on that cylinder was 8.2. A known good cylinder was 7.6. Spec for the engine is 7.3.]

In discussions of fire and fire prevention, it's common to talk about the required ingredients for starting fire: air, fuel, and an ignition source. Remove at least one, and you won't have a fire.

An four-stroke, internal-combustion engine (most car engines) is a little more complicated than that. I've heard people use the same three ingredients to talk about what can fail and make a car not run right, and that's useful, but it's an oversimplification if you want to understand what's going on inside the car.

The equivalent list for an internal-combustion engine are:
- air intake
- mix fuel with air (carburetion system)
- compress the mixture
- ignition to ignite the mixture
- a place for the exhaust to go

All of these items must be present for the engine to run at all. However necessary, the following relationships are also required to be somewhat correct for the engine to run:
- amount of fuel per air (called the "mixture")
- when the ignition spark happens in time compared to the peak of compression (called "timing" or sometimes "spark advance")
- amount of compression (called "compression ratio")

Further complicating matters, if any of the parameters changes rapidly, then there must be a mechanism to re-adjust the other parameters to keep the engine running. If this doensn't happen, then engine can stop.

When people think about running and adjusting engines, the mixture and the timing both at constant engine speed and as the engine changes speed are very important. If the mixture or the timing get far enough off, the engine quits and you know it. In the engine in my Beetle, the carburetor controlls the mixture and it has various mechanisms to set the mixture correctly and different engine speeds and loads. There is a vacuum port in the carburetor that's connected to the distributor; the distributor uses a combination of this vacuum signal and the engine rpm to set the spark timing.

I've spent a lot of time getting the distributor and the carburetor right in my engine. It's easy to tell when they're not working right; the car dies as I try to pull away from a stop sign or stumbles when I push on the gas or makes funny noises when I take my foot off the gas. When you get those adjustments wrong, it's easy to tell, because the car does something that you notice.

HOWEVER--there's a third adjustment in the above list. COMPRESSION RATIO (abbreviated CR). CR is the ratio of the inital to final volume of the fuel-air charge just before its ignited. It's a very important parameter for the running of the engine. The higher the CR, the more mechanical energy can be extracted from the fuel and so the more efficiently the engine will run. However, higher CR also means the engine will run HOTTER, which is very important in an air-cooled engine like mine.

It's easy to overlook CR because it's not something you can adjust from outside the engine, it's a function of the geometry of the insides of the cylinders. However, it's very important that it be set to a known, correct value for the engine to run correctly. Since CR is a function of the geometry of the cylinder/piston/head combination, it's physically possible that it's DIFFERENT for different cylinders (although that's almost certainly not what you want).

Compression ratio is measured by measuring the volume of the combustion chamber and the volume that the piston displaces in its motion, and plugging those into the compression ratio formula. The shape of the combustion chamber is very complicated, so it's measured with a liquid after using disk to form a top to the chamber with a small hole in the middle:


(This chamber isn't ready for volume measurement, obviously; one of the valves isn't installed. This is to illustrate how the plate fits over the chamber.)

I used water for the fluid, which isn't ideal, but it's readily available and easy to replace. Once I dry the chamber out with paper towels, here's my final drying method:



I measured the chamber volume of cylinder #1, which is the one that had been having problems, and cylinders #3 and #4, which had always been working right. One thing I discovered that the head on the right side of the engine had been machined so much that it had a ridge around the inside of the chamber:


that sticks up higher than the sealing surface (the shiny bit to the left of the ridge). My measurement of the volume of the combustion chamber for cylinder 1 was significantly different than for that of 3 and 4.

My best measurement for the CR of cylinder #4 is 7.6. My measurement for #1 is 8.2! By the way, the specification for the engine was 7.5 earlier in the 1600 engine, and 7.3 for the year of my engine.

So cylinder #1 has a CR of 8.2 instead of 7.3 (which means running hotter) and I also had air leaking into that side of the engine (means running leaner and thus hotter). I believe those two factors in combination are the cause of the stretching exhaust valve that has been my long-term problem with my car.

What to do about it? News at 11. :-D

pump it up

I got the engine out of my beetle a while back; I guess a month now. I've been busy with other things, but now I'm trying to get back to it.

To make sure my work bench could support the weight of the engine, I put on some diagonal bracing and built a "cradle" to distribute the weight to the main beams of the work bench. The cradle is the box that's sitting on the near end of the bench:



The engine's finally up in the air:


and ready to work on:



In addition to the cylinder head problem (which is the main reason I pulled the engine), the oil pressure in the engine has fallen below spec. This is partially because the internal parts are worn, but it's also possible that the oil pump itself:


is worn beyond what it should be. I've ordered another one; while I have the engine out I'll replace the pump with a known good one and see if that helps the oil pressure enough.

been busy

I got a bunch done on the beetle this last week and weekend. Too busy to blog, but I should be getting back to that. Here's a hint:

there was movement

I'm happy to say that the new battery I bought for the beetle from Sears does in fact fit in the battery compartment:


The new starter cable is installed, along with the new battery terminal, which has a plastic shield to prevent shorts from the seat springs.

And finally, the engine starts and runs after 6 months sitting still in the garage:



Now back to the brakes so I can actually drive it.

something old and something new

The October issue of Flying Magazine has an article about the new Cessna 162 Skycatcher. The Skycatcher is an aircraft specifically designed for pilots to train and fly using the new sport pilot rules for pilots and aircraft that have been enacted in the past few years. The idea of the sport pilot rules is to lower the bar for people wanting to become pilots, to get more people involved in aviation with less effort and money if their interests are more casual. The Cessna 162 is specifically targeted at that market. The article states that they will fly it for the first time in 2008 and plan to start delivering in 2009.

Interestingly, the engine chosen for the 162 is the very very venerable Continental O-200 engine. The O-200 is the powerplant of the Cessna 150, and in the years since the 150 went out of production, has also become staple of experimental aircraft, like the Q-200 tandem-wing. The January 2007 issue of Sport Aviation magazine has an article on the O-200 which states over 25,000 of those engines have been manufactured, and it has never gone out of production.

open wide and say ahhh

Over several more iterations of pressure testing, I was able to get consistent results. The only way to get completely repeatable results is to have the cylinder on the up-stroke. That seems to be the position where the rings seal the best (with the engine cold, of course). With the clyinder at TDC or in the down-stroke, sometimes something happens during that test that changes the results (the rings flip to the ascending state, maybe), but whether or not it does this depends on factors that I can figure out. Now that I have a consistent procedure, sometime I'll do the test again with the engine warm.

Over the past week I got the engine put back together, and yesterday I took time to do a few more things before getting ready to run test the engine:


I installed quick-release decklid hinge pins. For those thinking about doing this, it's very important to read the instructions. The quick-release pins are larger than the stock ones. You need to ream out the hinges with a 1/4" drill bit for it to fit. I found that wiggling the drill around after drilling the hole made the ends of the pass more open and the pins went in much easier.

I also went ahead and installed new hoses on the right side of the engine. That's standard 2-inch cold-air duct from the auto parts store with a spiral metal reinforcement. I got tired of the Aluminum foil ones not fitting and bending every time I touched them.

And I re-set the timing and then ran the engine and attempted to do a full tune-up procedure. Here's ready to test:


Setting up the carb was ultimately unsatisfactory. The engine definitely ran smoother, and the exhaust was noticably cooler than it has been. However, I think there's a chance I might have a leak in the intake system somewhere, which causes the engine to run erratically and makes it hard to adjust. I have some work to do today, but with a little luck I'll be able to test that this afternoon.

pressure 120 over 80

A gasoline engine is an Otto Cycle engine, which depends on being able to compress the air/fuel mixure and when that mixure has burned (and thus gotten hot) being able to mechanically harness the expansion of the hot gasses. The better the combustion chamber is at sealing, the better and more efficiently the engine will run.

So a very important diagnostic test of the health of a gasoline engine is to "pressure test" the cylinders. One way to do this is to employ a "leak-down" tester:


High pressure air comes in from the left, and the right side is connected to a hose with a special hose end that fits into the spark plug hole in the cylinder (instead of the spark plug). The gauge on the left reads the input pressure, the gauge on the right essentially tells you how "good" the cylinder is at holding that pressure. I recently discovered that the valves on the right side (cylinders #1 and #2) of my beetle had drifted drastically out of spec in roughly 1000 miles, which typically indicates trouble. I bought the above pressure tester on ebay and tested all four cylinders on Friday. The testing indicated the same story, cylinders 1 and 2 are not happy, cylinders 3 and 4 are fine. Here is the reading on one of the good cylinders:


So I will definitely need to pull the engine and at least look at the combustion chamber side of the right hand cylinder head to try to diagnose what's going on. Keep in mind, since I was just figuring out how to do this, I did tests tests with the engine cold, which is not the correct way. Now that I roughly know what I'm doing, I will take the car out for a drive and test the cylinders again with the engine warm.