Working on a CA with unit fully charged - it can be done.

One of the many virtues of methyl formate is that it is a liquid at atmospheric pressure and boils at 32 degrees C.
It had occurred to me for some time that in fact it should be possible to leave the methyl formate charge inside the cooling unit with the machine being worked on and open to the atmosphere.
There are some jobs which need to be done with the machine open to the atmosphere. For example:

• Rewire of the compressor,
• Repair of compressor to top plate welds,
• Repair of rust in cabinet top, and replacing evaporator bolts,
  
• Capillary tube installation.

Normally, the procedure would be to remove the refrigerant, and fill the machine with nitrogen before commencing work. However, the characteristics of methyl formate mean that this step can be avoided.

Why would you want to keep the existing methyl formate? In my part of the world, methyl formate costs $130 per litre, so it's rather expensive and wasteful to simply remove the original charge. As an example, I'd put a new charge in my CA-2 only one year before needing to open it up again for the capillary conversion.
I had considered recovering the charge, but due to it being partly vapourised, I imagined the difficulty of capturing all of it.
The more I thought about it, it just seemed so simple to just leave the methyl formate inside, in its liquid form, and do the necessary work.

Now, one of the characteristics of the CA is that when the machine is stopped and at room temperature, the whole thing is in a vacuum. So, in order to work on it open, it has to come up to atmospheric pressure.
When this happens, the methyl formate will revert to liquid and settle in the various parts of the system. As long as the temperature is below 32 degrees C, any vapour should be insignificant.

• So, point #1 is this must only be done on a cool day

How to bring up to atmospheric pressure? We don't want to just open the charge valve and let air in. That's bad for two reasons: 1) air contains moisture, 2)air contains oxygen which will support combustion of the methyl formate. Remember, methyl formate is flammable.

• Nitrogen is used to bring the system up to atmospheric pressure. It will not support combustion of the methyl formate, and it will not contaminate the system.

How to know when the cooling unit is at atmospheric pressure? A problem here is that the nitrogen regulator won't really give an indication. If, for example, you set it to say 10 psi, the regulator gauge will show erroneous readings because of the regulator working into a vacuum. So, a separate gauge is used, connected to the charge valve. Between the regulated nitrogen supply and the other gauge is placed a valve. If using a valve set as shown in the photo, this already has such a valve. From time to time, the valve is shut and the cooling unit pressure observed. When it is at 0psig, or a couple of psi above, it's ready. Disconnect everything, and slowly open to atmosphere. Do this very slowly so as not to stir up the methyl formate.

Before work, the cooling unit should be placed on its side, with knobs facing down. Place padding underneath, so the condenser is not dented and the knobs are not resting on the surface on which the machine is laid down. In this position, pipes to be cut will be horizontal, minimising shaving entry, and liquid methyl formate will settle away from the openings.

We're now ready to start work. There'll undoubtedly be some vapour in the system, so don't take to it with a blow torch at this point. Instead, cut the required lines, using a tube cutter where possible. If you have to use a hacksaw, have the pipes horizontal so the shavings settle near the cut, where they can be brushed out, rather than falling inside the system.

As methyl formate is a solvent, it evaporates fairly quickly, so the severed pipes will appear dry and there won't be any odour. That's how we want it.

Use a stove lighter or a long match to place a flame at the pipe openings - with the openings pointed away! There should be no ignition, if all has gone to plan.

If you have to leave the unit unattended for a while, seal the openings with plastic caps or similar to prevent evaporation.

When doing a capillary conversion, make up the reducer, filter, and capillary tube as a sub assembly first. Thus, you only have to apply the blow torch to two places on the cooling unit. It also allows the use of nitrogen while soldering, which won't be possible if making up the unit on the fridge...

When it comes time to solder everything up, one has to go against the rules. This time, we don't want any nitrogen flow. As I discovered, this will stir up the liquid methyl formate, and it will also force some vapour through the system - which is probably not a good thing with flames nearby.

The CA is forgiving with contaminants, and the copper oxide generated during soldering should not cause a problem. (A filter is essential when doing a capillary conversion in view of this, and is a good idea anyway).

Soldering up is done in the normal way. Make sure you can't smell any methyl formate and the pipes are dry before starting. Again, keep away from the connections just in case it does burst into flame.

Normally, at this point we'd vacuum the unit and refill it. But, we don't want to vacuum it now for the obvious reason that the methyl formate would start to vapourise and be sucked out. Instead, we let the compressor do the job. In fact, it is exactly the same procedure as purging for NCG's. All the nitrogen and air must be removed and the vacuum restored.

Simply follow the purge procedure; oil in the charge valve, 3 minutes cracked open, 1 minute closed, and repeat. (See the GE training film excerpts for detailed instructions). Because the unit is initially at atmospheric pressure, it takes a good deal longer than normal. For my CA-2 capillary conversion work, it took about 1/2 hour before the top of the condenser started to warm up. Another 1/2 hour and the float chamber was warm. Another 1/4 hour and it appeared to be sufficiently free of NCG's.

This was also a useful insight into seeing what happens with NCG's and a capillary conversion. With no float held shut by NCG's in the float chamber, what is the effect? There's no rattling, and as far as I can tell the effect is simply that the evaporator doesn't run as cold as normal. The frost line appears normal (what frost there is). As with the float valve equipped machine, purging is done until the float chamber is warm as the condenser, and in the case of the capillary equipped machine, the filter is warm.

*Disclaimer: This article is a description of the method I have used to work on a machine charged with methyl formate. If you do not feel confident handling methyl formate, don't do it. Read the MSDS and know the properties if you do. Think about what you are doing. I can't accept responsibility for anyone's injury if the procedure is done differently, or if any other unforseen circumstances should occur.

And finally, I'd like to see the look on the guy's face who wrote this quote as I applied the blowtorch to the fully charged machine..."I have more fear of methyl formate than I do of gasoline".