View Full Version : Oil Concentration Sensor

Erik Detroit
29-10-2005, 04:13 AM
Hi guys,
I'm an automotive refrigeration engineer (in Detroit) and I've been working for some time on an oil concentration sensor. I'm not sure what this is called in your various refrigeration fields, in automotive it's usually called the Oil In Circulation (OIC) expressed as a percentage. The meaning is mass of oil circulating per mass of total (oil + refrigerant). I've never liked that term, so I'm calling it "Oil Concentration".

The sensor is not what I would call ready to market yet, I'm still working on some calibration difficulties. Before I quit my day job, I'd like to know how big the potential market is. In other words: Is there any use for such a thing?

The nature of the device is rather expensive (for discussion let's say 7000 USD), and the calibration procedure is time consuming. The benefit of course is an instant (and continuous) indication of the actual oil concentration in the liquid line. The sensor could be quickly attached to a service fitting to get a single reading, or the sensor could remain attached and connected to a datalogger.

The automotive application would be the latter, an OEM or a tier 1 system supplier would connect the sensor to the refrigeration system, and along with all of the other validation of the refrigeration system they would have data to show that the compressor is (well, poorly, borderline, not at all) lubricated in each operating condition.

Please do me the favor of telling me if you think there might be a demand for this thing.


30-10-2005, 08:21 PM
Hi Erik

Do you also have a chart to compare the actual readings with "high/low" conditions. How would the sensor interpret the readings during part load run times when oil circulation is different to high load conditions?
Will the equipment be suitable for all types of refrigeration systems?

Erik Detroit
01-11-2005, 03:33 AM
Hi Frank,
The measurement of oil concentration is not being performed in my industry other than at a few specified conditions. The mindset to monitor the oil concentration and adjust the oil quantity in a system does not exist in the automotive industry. Rather, rules of thumb and generalizations are used such as at a given set of conditions, the concentration should be X.X% or greater. These rules of thumb are generally very conservative to avoid an increase in warranty costs. The need for this sensor is variable displacement compressors which are becoming extremely popular in automotive, much like variable speed for stationary. The rules of thumb are out the window, since they were formed with experience of fixed displacement clutch cycling compressors.

Getting back to your point...
If you had an oil concentration sensor, and you could gather data of successful systems, I think you could make comparative data as you suggest. The sensor itself is very much a quantitative instrument. The output is only in units of (mass oil)/(mass refrigerant), or of course oil% of the total.

Your question about part load situations is really what has driven the desire for such a device. Partial loads deliver lower suction line gas velocities in a pipe that is sized for full load. The lower the velocity, the slower the oil is pushed along the pipe. If the flow is slower, there is more being stored and building up in the suction line, and possibly the superheated passes of the evap. To answer your question directly, there would be no difference in the operation or interpretation of the sensor. The resulting information in all operating conditions would be only the mass fraction of oil present. At low loads, the sensor would take more time to respond to changes in the concentration than at high loads.

The property I am measuring gives very good resolution and repeatability with R-134a and automotive PAG oils, and this is the only combination (refrigerant/oil) I have calibrated for. The property depends greatly upon the refigerant used and the oil used, and I would have to characterize the property for each refrigerant/oil combination.

I'm not sure if your question was about different refrigerants or different cycles. (absorption?) The sensor would work for vapor compression systems in which the oil is miscible in the liquid refrigerant.

Thanks for your thoughts,