The preferred sampling point is at the oil piping going into the priority valve housing, from the factory a quick disconnect fitting is installed there, usually referred to as a "vampire" fitting. CAT SOS sample kits have tubing adapters to interface with those sample ports.

Live sampling is the preferred method for the larger engines. An alternate method is to use a suction pump and tube thru the dipstick tube into the sump, however this method tends to provide inconsistant results.

Here is a link to the types of CAT oil sampling systems,
https://parts.cat.com/en/catcorp/fluid-sampling-eq...;

Hope that helps, MikeL.

You could also take your sample when you change your oil. But, make sure you let oil flow for a minute or so before you take your sample to
avoid getting sediments from the bottom of your pan.

That's what I do and IMHO its the right thing to do.

Does anyone have any cons to this method?

Between the oil pump and oil filter. This is especially true if you're running absolute rated filters. I have some gearboxes that periodically start clogging filters with metal but there samples are always clean because nothing makes it past those filters. Another thing to keep in mind is that oil samples don't catch catastrophic component failures. If you can see the particle in the oil you won't see it in the sample result.


On a side note, oil samples are cheap and help operators comply with ISO or whatever other supervision your customers want. They are otherwise completely worthless unless you pay for an upgraded tests. Coolant samples are FAR more valuable from an operator's standpoint. Over my last 10 years I have overseen 1.5 million hours of engine operation and have never seen a lubrication related failure. Cooling system problems on the other hand...

What I found was the following.

Sample from live fluid zones.
sample from turbulent zones such as elbows
sample downstream of bearings, gears, pumps, cylylinders and actuators.
Sample machine during typical work conditions.
Sample from the same sampling point for any particular piece of equipment.
Sample after flusing a small quantity of oil (0.5-1.0 l) through the sampling point - and witout operating the sampling valve between flushing and sampling.

TUGBOATENG excelent feedback I am going to take into account.

TUGBOATENG is seems that the in oil analysis we could not find anything, and maybe we are doing something wrong.
I upload more information about this. Thank you for all replies.

I attach a scheme of the system (it is smimilar that we have in plant).

• https://files.engineering.com/getfile.aspx?folder=87bb806b-14e3-42da-bbfe-5

I've attached on older CAT D3600 Application and Installation Guide for reference, I prefer to still use the older publication since it only dealt with the 3600 series engine, the newer guides apply to multiple engines and applications.

On page 206 of the attached is a description of the lube oil system, and on page 221 is a diagram of a 3608 lube circuit, 3606 lube circuit is essentially identical, just two fewer cylinders.

Older engines had the factory supplied oil sampling port in the pipe at the oil pump outlet going to the oil coolers, when the G3600 engine (gas engine) came out, the factory moved the sample port to the piping going to the inlet of the priority valve.

Oil sampling for wear metals targets metal particles in a specific size range, if the metal particles are too large, they don't properly process thru the analyzers. So along with oil analysis, regular inspection of the oil filter media and regular maintenance of the oil suction screen is also important.

Here is a link to a resource for excellent info on oil analysis, https://www.machinerylubrication.com/Read/30976/oi.... There are a lot of links at that page and I think you will find a lot of good info. If you really want to expand your knowledge their book the "Wear Particle Atlas" is an excellent resource, I used it as class material in a class last year and had positive feedback about the material, many engine and equipment manufacturers provided lots of the info in that guide.

The engine oil filters typically don't remove wear metal particles, the standard CAT (and most aftermarket) 3600 oil filters are nominal 20 micron, but your centrifugal filters can remove wear metal particles from the oil stream, so make sure you are looking at the debris coming out of those filters as well, I found the paper liners to be a big help in holding the debris together long enough to get it out on a table and take a close look at it.

Lube oil analysis essentially covers four major areas, wear metals (indications of engine wear), lube oil base stock condition, lube oil additive condition, and contamination. A properly done analysis gives you a lot of info, a poorly done one is worse than worthless because many times it sends you on a wild goose chase. For a good analysis you also need to make sure the lab has a sample of your new oil for proper comparison. If you store oil on site you should also sample your new oil storage as well at regular intervals, usually right after a new oil delivery, that's when all the crap in the storage tank gets churned up and sometimes you may find you're not getting the lube oil you asked for.

So, for your engines preferred sampling method is at regular hourly intervals (250 hrs was the CAT recommended interval), from a "live" location with the oil at normal operating temperature, make sure you also know the total number of hours on the oil, and how much makeup oil you are using if you want best results.

Unfortunately my experiences are much different than TugBoatEng's, I dealt with a number of engine failures and issues due to lube oil related problems, but it was part of my job at the time, and I learned a lot along the way.

Hope that helps, MikeL.

• https://files.engineering.com/getfile.aspx?folder=69cf644a-1c01-4ac4-9206-d

When taking a oil sample the first thing you want is it to be representative of the oil flowing through the engine. If no typical oil sampling point is available on the engine, the easiest way is to take a sample from the oil reservoir. Consistency in taking the sample is the most important thing: under the same conditions every time and any time. Not from the bottom of the reservoir (where there may be excess dirt), but always on the same spot at a certain distance from the bottom. Where you take the sample is less important then how you do it - the aim is to get consistent readings.

Once you have a number of samples you might be able to discern a pattern - the longer the oil is in use the more wear metals and soot you will find, and viscosity may increase due to oxidation or presence of water or may get thinner due to fuel dilution etc.

In most cases the laboratory that reports wear metals will do so in ppm. The data obtained after a longer period will show more ppm and that may seem cause for concern. It not necessarily is. The number of ppm that is reported is the ppm that was left in the engine oil at the time of sampling - the figure is influenced by the time the oil was in use, the amount of oil added and the amount of wear particles that was taken out by filters and thus it not necessarily represents the amount of wear generated. To put it somewhat bluntly: if you mount a better filter, the amount of wear particles measured will be lower - but not necessarily the actual wear....the better filter will most likely "catch" more wear particles but they still can be generated.

What you really should be looking for is the relative amount of wear over time. A engine in good condition will produce a certain amount of wear particles over time at a more or less constant rate that may vary with the engine load. It thus is more useful to not just look at the amount of wear in the sample but relating it to a standard numbers of hours such as ppm/500 hrs and the increase of the number of ppm over the last 500 hrs compared to previous samples. That way you can more easily detect changes in the wear rate which may indicate some form of malfunctioning.

You should be somewhat careful using the information given by the engine builder. Those figures are rather global and thus not necessarily applicable to your engine under it's operating conditions. After a number of samples you will be able to develop somewhat more precise rejecting limits for your particular engines than the more general ones as supplied by the engine builder. You may also notice that different makes and types of engine oil will generate different wear figures.