Convert flooded R22 PHE chiller to NH3?

[garyb]

We are considering converting an existing flooded R22 glycol chiller to ammonia. System has two VMY336 screws with liquid injection cooling, secondary oil separation vessel, evap condenser, receiver, thermosiphon PHE and surge drum. System is 95% steel so only a little bit of pipe work to replace, shaft seal will need replacement and surge drum oil recovery will need to be reconfigured - all pretty straight forward. System charge is around 600kg with mineral oil.

The big question is how clean do we need to get the system so that we are not going to have R22/NH3 contamination issues? We are well aware of the toxic and corrosive compounds that this can create. No matter what we do we will never get all the old oil out of the lines and I imagine that the R22 has been absorbed into the pores of the inside of the steel pipe work. Will multiple evacuations and dry nitrogen breaks sufficiently remove the residual R22? I don't believe that flushing is an option as getting all of the flushing agent out will as much of a problem in itself.

Has anyone done an R22 to NH3 conversion successfully?
Will multiple evacuations and multiple oil changes after restarting be adequate?

[Yuri B.]

Hello Garyb.
Evidently too unique an undertaking for engineers of this forum . You should probably consult with comp's manufacturers about compr's compatibility with ammonia.

[RANGER1]

garyb ,
If you evacuate properly with dry nitrogen breaks as you mention it will be ok . Have been involved in a few large changeover with no problems at all ( oil MOBIL SHC226 , didn't change) .
PHE 'O' ring material would have to be checked . We didnt bother with shaft seals or anything unless you know of incompatability . Compressor 'O' rings etc should be ok as well as when you buy a overhaul kit its usually compatible for ***** and ammonia .
Maybe surge drum/ t/syphon height be checked for ammonia duty .
One large system i know of was'nt evacuated properly and a chemical reaction occured causing a type of powder to continually block liquid line strainers for a few days , but it did recover and all was ok ( oil suniso 4GS , did'nt change) .
Obviously all brass / copper fittings have to be changed to steel .
Oil drain points and collection points would also have to be installed around surge drum etc

[US Iceman]

The two biggest issues I am aware of were mentioned by Ranger1.

One large system i know of was'nt evacuated properly and a chemical reaction occured causing a type of powder to continually block liquid line strainers for a few days

This is the problem issue if it happens to you.