cars

A Cool Trick for Working On AC Systems

~ joelshaw@sbcglobal.net ~ Leave a comment

Last week was Mechanic Week for me. I knew I was going to put a new clutch in my wife’s little car, but I was also provided the surprise opportunity to replace the locked up AC compressor on Red Truck. With about 200k miles, I am kind of proud that it made it this long.

Everything is really easy to get to and the truck has high mileage, so I decided to do it right: new compressor, flush everything, new dryer, new orifice tube, pull vacuum for an hour, etc. With everything out in the open, I figured I might as well. I was able to replace the compressor and dryer in about 30 minutes. Then I got the orifice tube.

If you are not familiar with AC systems, the orifice tube is a small nozzle that allows the freon to expand from liquid to gas. Cooled gaseous freon subsequently goes through a heat exchanger to cool the car. The orifice tube is a small 1-piece device that simply slides into the high side freon tube before the evaporator. The orifice tube has an integral tang at the top for easy removal. Pull on the integral tang, and the whole device slides out of the freon tube. It would make sense to place that tang near the opening of said freon tube. Ford thought otherwise (shown to the right).

It turns out that needle nose pliers can be modified to reach the tang of the orifice tube. Modified needle nose pliers are very good for un-integral-ing the integral tang from the 1-piece orifice tube, leaving nothing to grab. Luckily, Al Gore loaned me his internet so that I could find a solution.

This seems to be a fairly common problem. I found that the “old guys” get a long wood screw, wrap tape around it as a guide, screw it into the top of the orifice tube, and yank on it with unmodified pliers and all of their might. Despite Ford’s best efforts, I had it out in minutes. Does this mean I am now an “old guy”?

While I am talking about Ford A/C systems, I would like to complain about something else they did. Some engineer somewhere decided to use a rubber ball instead of a shrader valve on their high side service ports. It might have saved a penny at some point, but I’ve never met one that didn’t leak. I’ve also met very few high side couplings that will reliably press the rubber ball enough to get a consistent reading. Why, Ford? Why?

Pulling all of this together, I wish someone would take Ford’s high side service port design and shove it up the engineer’s orifice tube. Also, try using a wood screw and some tape to remove your orifice tube if you (or it) are in a bind.

 

Needless to say, the gallant steed is all back together and on the road. Hopefully the AC system in Red Truck will keep together for another 200k miles.

How I Found Myself with a Foundry

~ joelshaw@sbcglobal.net ~ 5 Comments

Ever wanted to know the steps to building and using your own foundry? It has been quite a few years and involved some jackassery, but as I recall it went something like this:

 

  1. Start working on a Lotus Esprit and realize you need a unique washer that is hard to come by (about $5 if available).
  2. Buy a metalworking lathe to make said spacer.
  3. Discover that the lathe is missing a part.
  4. Get frustrated that the part is just an aluminum block but costs $100 with shipping.
  5. Realize that the part wouldn’t be too hard to cast.
  6. Source aluminum flashing, ceramic wool, ceramic paper, sodium silicate, nichrome wire, a temperature controller, a crucible, and castable refractory mortar (totaling well over $100).
  7. Assemble sourced items to resemble a foundry.
  8. Obtain approximately 100 cans of Miller Lite.
  9. Drink approximately 100 cans of Miller Lite.
  10. Melt approximately 100 cans of Miller Lite.
  11. Sand cast the part for the lathe.
  12. Use the lathe to make the $5 washer for the Esprit.

Seems legit, right? I have a kiln now and will probably never use the foundry again, but I am glad I built it for two main reasons: it was a lot of fun and it was really dangerous!

I am including some pictures, but I am not going to include detailed instructions on actually making it. It is unsafe, and doing anything like this is a very bad idea. However, I will include some generalities and observations in case anyone wants to use any of this for a more rational project.

  • A good source of nichrome wire is from the heater elements of an electric dryer, and it seems to be good to about 2000 F. It is important to do your homework on the length. It must have enough resistance to limit the current or it will be annihilated as soon as it is powered up.
  • Sodium Silicate (or water glass) is really neat stuff. I used it as a stiffener and makeshift adhesive for the ceramic paper, but it can also be used for making things fire resistant.
  • Ceramic wool seems to be good to at least a couple thousand degrees Fahrenheit. It is also a good thermal insulator. I had no problem touching the outside when it was 1700 F inside. You can also see that the electrical tape around the flashing didn’t melt or deform after operation.
  • I heard that ceramic wool became readily available due to technology development for the space industry. That might be a rumor, but I would rather perpetuate it than research it. Feel free to let me know if I am wrong, and you will be eligible for today’s Pettiness Award.
  • The crucible is not shown, but I initially used a cast iron pot to hold molten aluminum. I do not know if anything leached from the cast iron into the aluminum, but it hasn’t effected the functionality of the part.
  • the whole thing ended up fairly light weight. I remember it being about 20 lbs.
  • If you are thinking about making something like this to avoid buying a part, don’t. Just buy the part.

Feel free to comment if there are any questions. I don’t think I will ever use this again, but I might use some of things I learned for another project. Anyone interested in a portable electric “wood fired” 1000 Fahrenheit pizza oven?

Next Boost Test

~ joelshaw@sbcglobal.net ~ 1 Comment

KIMG0048For those of you on the edges of your seats about the functionality of the cheap ebay boost controller, you can relax. It didn’t work. It would initially over boost before settling back down.  I took apart the tee, and found this (photo on left). I thought the tee contained just a check valve, but the spring was strong enough that it acts as a relief valve. It would boost up to about 18 psi, crack open the ball, and then settle back down. I dropped the spring seconds after taking this picture, and it bounced off the shop floor and into oblivion. I don’t think that is necessarily a bad thing. I will explain.

 

The following three sketches show:

  • The stock configuration
  • What I thought I was buying
  • What I actually bought

The first uses an electric solenoid to control boost (which the Grand National tries to do blindly). The second uses a “Grainger Type” boost valve that is so named because one can modify parts from Grainger to make it.  I didn’t initially go this route because adjustment results in burned knuckles.  It consists of a simple adjustable home made relief valve. The third sketch shows the Cheap Ebay Manual Boost Controller (CEMBC) that I purchased. As shown in my last post, the CEMBC came with a tee and a variable orifice. Internal to the tee is a fixed orifice and relief valve. The problem was that the boost required to overcome the relief valve was greater than the boost I want to dial in. It was overboosting before the relief valve even opened.KIMG0049

Now for the next test. I’ve completely removed the ball and spring in what I call the Modified Cheap Ebay Manual Boost Controller (MCEMBC). I think that this will provide more accurate feedback to the waste gate actuator, and I do not think they are necessary if no vacuum actuated circuits are connected to that loop. I initially thought that this would cause inconsistencies between immediate Wide Open Throttle and rolling into the throttle more slowly. I’ve thought about it more, and now I think that its operation will be similar to a pneumatic version of a voltage divider. The signal to the waste gate actuator should be a nice, smooth percentage of the boost pressure.KIMG0050The only disadvantage I see to this setup is increased spool up time because it won’t lock the waste gate closed until it reaches the desired boost. Instead, it will open gradually. I think my final solution will be a combination of a “Grainger Type” controller and the MCEMBC. It will really depend on the adjust-ability of the fuel map when I get a chip I can tune.

This post really isn’t for the general populace. I just want to log what I am doing so that I can bounce it off of other car guys to see if they agree. I promise to make the next post about cats, chickens, interior decorating, or something along those lines.

Getting Closer

~ joelshaw@sbcglobal.net ~ Leave a comment

My KIMG0044Grand National is getting closer to running correctly in its almost stock form. I put a manual boost controller on today. Before you start complaining about everything being zip tied together, I need to point out a couple of things. First of all, I find the term “zip tie” rather offensive. They are called racing ties. Secondly, this isn’t permanent. Even though they seem to have decent reviews, I don’t trust the boost controller I bought.  I will re plumb it in a semi professional manner if it works as advertised.

The biggest reason that I don’t trust the boost controller is that it cost only $11.99 on ebay (with free shipping). The second biggest reason is its theory of operation. It simply has a variable KIMG0043orifice that releases boost feedback on the waste gate side of a combination orifice/check valve. It seems to me that it will change the ultimate boost if I am at Wide Open Throttle vs accelerating at a moderate pace. I pondered tossing this out when I figured out how it works. Then I noticed the smaller printing. It states “HIGH PERFORMANCE” right on the label, so I decided to give it a try. I hope to report back with good news.

Not Enough Power

~ joelshaw@sbcglobal.net ~ 2 Comments

So no luck with the chicken coop.  The solar panel I was using was labeled as 5 Watts. However, it only put out about 20 mA (.24 Watts) in direct sunlight. That wasn’t even enough to run the power supply.  I assumed that it would put out more than that in the shade. I guess I need to buy a better panel.  More money, and I don’t even have any chickens yet.

 

2015-08-04 001

On another note, I’ve been struggling with sorting a Buick Grand National for some time now. Today, I figured out that the previous owner had installed an adjustable boost actuator at over 20 psi.  I never noticed how high it was set because I had a big crack in one of the headers. I welded that up last week, and it started blowing off vacuum/boost lines. I guess someone’s solution to low boost was to crank on the actuator rod rather than fixing the problem. I used my air compressor and a pressure regulator to open the waste gate (pictured), and it didn’t fully open until almost 25 psi!  I cranked it back down, and I have a Manual Boost Controller on order.  I can’t wait!  Once I get it tuned back down and running predictably, I can start cranking it back up and spinning the tires!