Insane Hydraulics
Bold and audacious blog about fluid power
I want to share my first experience with a "reasonably advanced" filter cart from RMF. I use simple filter carts all the time, but this was the first time I got to use a more advanced model, equipped with a contamination monitoring system. Unfortunately, I don't have any hands-on experience with other brands. It would be great to make a direct and unbiased comparison (a competition of sorts), but these babies come at several Keur a pop, so, yeah... Maybe some other time...
Anyhow, we got this brand new RMF OLUS1B60G1BA20010M offline filtering unit. It's a nice thing to have around a respectable shop, don't you agree?:
I like the way it looks (now that I said it - I guess it looks like any other filter cart in the world). Still - the specs are great. OLUS stands for "Off-Line Unit Smart", and it does two things:
What is particularly interesting about this sensor is the fact that it promises the accuracy of ± 1/2 code for 4, 6, 14 μm (c), and ± 1 code for larger sizes (at least according to the respective web-page, even though the brochure has a "3" in front of the "1/2" for some reason, so this may as well read as 3.5 codes, but you know - catalogs and brochures are always filled with mistakes and typos).
By the way, about the brochure - I especially like the phrase "...to be mounted ... where space and costs are limited...", which sounds silly to me because this is a €3000 piece of monitoring gear.
So, in theory, these numbers are great, but how about practice? As soon as I received the cart, I immediately hooked it up to our small test bench to test it out (and to filter the oil as well, naturally). I also connected our trusty Hydac FCU 2210-1, which was calibrated about four months ago (and I have a nice shiny certificate to prove it), to see how the so-well-advertised contamination monitoring sensor would fair:
After spending a day "toying" with all this sophisticated gear, I can tell you that I was genuinely surprised by how inaccurate the contamination monitoring sensor turned out to be. I ran it with cold oil, and hot oil, took samples in different places of the system, made sure no air bubbles we present, and tried it all - and I would always get a discrepancy of several codes between the Hydac FCU and the CMS. Just have a look at one of the logs I made (only a small portion of it):
| Time | Hydac | RMF | 4 μm dif | 6 μm dif | 14 μm dif |
| 14:36 | 16/10/7 | 21/16/9 | 5 | 6 | 2 |
| 14:46 | 16/11/7 | 21/15/7 | 5 | 4 | 0 |
| 14:54 | 15/10/7 | 21/14/7 | 6 | 4 | 0 |
| 15:05 | 14/10/7 | 20/13/7 | 6 | 3 | 0 |
| 15:16 | 14/10/7 | 20/12/8 | 6 | 2 | 1 |
| 15:32 | 13/〈10/〈7 | 19/11/6 | 6 | 1 | 1 |
That is about five to six codes difference for small particles. This is a lot. Too much, actually. Code 21 stands for 10000 - 20000 particles and 16 for 320 - 640 (per one ml). If we, for example, look at the Rexroth recommendations, published in their very nice booklet on oil cleanliness, we will see that for a system that runs variable displacement pumps, the first line from the table above would mean "perfectly acceptable" from the Hydac's point of view, and "stop everything or you'll ruin the pumps!" from the RMF's.
I actually contacted RMF with this issue, but so far only got a vague confirmation that "this is strange". Maybe I'll get an answer someday, although my experience with large brands tells me that most likely it'll stop here. I'll definitely post an update though if this happens.
The device does have a registered calibration date from 2021, so maybe re-calibration could solve this issue?
But then again it's a brand-new machine that is supposed to work fine out of the box. I am not sure what's going on at this point, but I will stick to my Hydac for doing the monitoring and the RMF for doing the filtering, at least for now.
One thing is for sure - for bigger particle sizes the readings were not that much off, and the relative evolution of the oil cleanliness still can be perfectly monitored even if the readings are offset. Unfortunately, I can't trust absolute values. It's a shame.
You can see that for a relatively small oil tank (200 liters) a several-hour-long offline filtration made quite a difference, especially if I tell you that when I started I got the 20/16/10 from the Hydac (23/19/11 from RMF).
I also found it interesting that due to the relatively low oil flow, most of the oil seems to be passing through the bottom of the filter element - at least that is what I see in thermal images:
I guess as the part of the filter element most of the oil passes through gets loaded with particles, the "filtering" will gradually go up the element. There's a tube inside the element that channels the oil up to the top cap (this is why it doesn't have an air bleed, btw), but still, only the bottom half of the aluminum body heats up.
My first impressions then:
a) the filter cart, obviously, works. And why shouldn't it? When you circulate oil through a high-efficiency element, you capture particles.
b) The contamination monitoring sensor also works, in the sense that it gives you a reading that you can compare to a later reading. You can use this to monitor the evolution of the particulate content, but you can't use it as an accurate instrument. Such a shame, in my opinion. Especially if you take into consideration the price point of this gadget.