Discrete Scroll Compressor Model

[gmansparky]

Hello,

I am new to this forum and this is my first post err.....well rather question...(I am sure I will have more qns than answers!!!!!!!).

I am trying to build (software / simulation) a controller for controlling the Refrigerant pressure. The actuator is a Motorized Ball Valve. I need a very simple representative model for the process in particular I am interested in getting typical discharge Pressure (controlled variable) values on the outlet of the compressor.

The compressor being used is a Discrete Scroll compressor I have been looking for a simplified model of a discrete scroll comp. but haven't been successful in finding anything concrete. I understand that the Discrete Scroll varies load by turning a solenoid valve ON / OFF which results in a Pulse Width Modulated loading pattern (duty cycle dependent), I am thinking this would introduce Non Linearity into the system that I need to contend with (from a controller design perspective).
It would be of great help if you can help me with this.

Many thanks in advance for your help. :)

Gman.

[desA]

Take a look at the Copeland Digital Scroll website - this may help.

I would wonder if much of the non-linearity issue would not perhaps be damped out by the refrigerant vapour compressibility?

Would be interesting to model this... Sounds like an extremely interesting project. Which simulation environment are you working in?

[gmansparky]

Hey desA,

Thanks for the information yes I did already take a look at Copeland's Discrete Scroll Compressor website and gleaned whatever little knowledge I possess about these compressors.

I propose to simulate the behavior in Matlab / Simulink environment which is quite powerful. So you think the energy expended during compression would damp the Non Linearlity?.....I hadn't thought about that at all...what effect would the switching have on the Load Distribution?

Here's another question what models exist for other compressor types ?(excluding the discrete scroll). Would these models enable computation of Comp. Discharge Pressure?. If yes what is a good place to look for them?

Thanks again, I appreciate your help.

Gman

[Peter_1]

You also could use a VFD controlled compressor (piston, scroll, ..) or an unloader on a piston. You then have within certain limits a stable flow.

The pulsations of the digital scroll are indeed flattened out while running due to compressibilty, line resistance,...
If it wasn't like that, they shouldn't be used in AC's.

What do you mean with "what effect would the switching have on the Load Distribution?" The on/off switching is to adapt the needed load to the provided load.

"Controlling Refrigerant pressure" do you mean refrigerant or condensation pressure?
Why do you want to maintain anyhow refrigerant pressure? If for energy saving purposes, then you're on the wrong track.

[desA]

I propose to simulate the behavior in Matlab / Simulink environment which is quite powerful.

Matlab & Simulink are nice simulation environments - very intuitive.


Here's another question what models exist for other compressor types ?(excluding the discrete scroll). Would these models enable computation of Comp. Discharge Pressure?. If yes what is a good place to look for them?


Have you tried EES? You could possibly try out their demo version to see if it fits in with your style.

[gmansparky]

Hey Peter,

Yes the pulsations will be flattened out on an average i.e. load matching will take place. However there seem to be pressure spikes (instantaneous not long run average)created due to this Modulation process. I am interested in reducing these pressure spikes and keeping Refrigerant pressure stable. These spikes cause considerable wear / energy loss especially under dynamic loading conditions.

I mean the refrigerant pressure on the exit side of the Heat Exchanger (condenser).

Thanks.

Gman

[gmansparky]

Hey Des,

Yeah Matlab / Simulink is pretty untuitive. I've never used EES will test drive the Demo version.

Thanks.

gman

[desA]

Hey Peter,

Yes the pulsations will be flattened out on an average i.e. load matching will take place. However there seem to be pressure spikes (instantaneous not long run average)created due to this Modulation process. I am interested in reducing these pressure spikes and keeping Refrigerant pressure stable. These spikes cause considerable wear / energy loss especially under dynamic loading conditions.

I mean the refrigerant pressure on the exit side of the Heat Exchanger (condenser).

Thanks.

Gman

Pressure pulses exist in a standard scroll compressor discharge line - actually, all the way through to the liquid line & beyond. This can be observed on a standard pressure gauge set. On many systems that have integral pressure gauges, they are heavily damped.

Practically, pressure fluctuations will exist from compressor discharge all the way to evap inlet - these compressors are cyclic in nature - not continuous. The main thing that would be of importance in your analysis would be the energy of the discrete scroll pulse, against the existing (standard) characteristic.

[R410engineer]

Hii everybody...
I m new to this post..
Im also working on simulating Scroll compressor
I want to ask that anybody successfully modelled scroll..???

Cheers

[Tesla]

Hi Guys
I gather the objective here is to smooth out those fluctuating pulses, taking out the sharp spike. I would suggest as opposed to simulation perhaps measurement and analyze the result. Then use proven formula to calculate the correct response.
This method would take into account any unforeseen variables - such as (perhaps) capacitance of condenser, reflection of pressure pulses, resonance, Schumann effect, second order lag etc. Over a range of varying conditions of COMPRESSION RATIOS (in particular) for bias and pressure range.
There are some pretty flash pressure sensors on a google search like tourmaline optical which can measure into 100kHz.
I may be on the wrong track here - so please inform me if so. The confusion in my mind is how could a ball valve act so quickly. And if it could - it could close off on the end of the down slope to prevent discharge (some) bypassing back into compression chamber (particularly for recips not so much scrolls) thereby increasing efficiency. Interesting thought anyway.