Comparison of the Turbo / Intercooler Glassware Test (Using DIN 51535) to a Fired Heavy Duty Diesel Engine Test 2004-01-1957
The turbocharger/intercooler (TC/IC) glassware test (using DIN 51535) was originally developed to mimic the effect of lubricant oil misting on turbocharger and intercooler performance in large stationary diesel engines. Oil mists are generated in the sump and pass into the turbocharger and intercooler via the closed crankcase ventilation system. One of the potential negative impacts of this oil mist is degradation products being formed in the high temperature turbo parts and housing, in some cases leading to turbo boost pressure loss (TBPL). The test uses simple laboratory glassware equipment to mimic oil mist generation and a furnace tube to represent the hot turbocharger parts, and is used as part of selected OEM specifications.
Statistical analyses using the turbocharger results from the Daimler-Chrysler Euro II OM441LA engine test and the TC/IC glassware test has shown that the TC/IC glassware test does not accurately predict either turbo weight change or TBPL in the OM441LA and the weight change of the turbo in the OM441LA does not predict the TBPL accurately either. This can be illustrated by the fact that:
An oil showing excellent performance in TBPL (-4.5 to 2.5%) can have a TC/IC deposit mass ranging from 50 to 190mg (pass/fail limit is 120mg);
An excellently performing oil in turbo weight change (< 5g) can have a TC/IC deposit mass from 50 to 190mg also;
At 0% TBPL, the weight change of the turbocharger assembly can be high (< 60g);
At high TBPL (20%) the turbocharger assembly weight change can be very small.
Experimental work is described using a standard TC/IC glassware test, in which the heat input to the furnace which maintains the tube temperature is not operating. This was to investigate the composition of the oil mist, from which an understanding of the nature of the subsequent deposits could be developed. Results from this work show that the test generates an oil mist which is composed primarily of bulk oil and is not strongly biased by volatilisation effects.
Based on the data described in this paper, it is recommended that the TC/IC test should not be considered as a good predictor of TBPL in CCV HDD engines.