centrifuge to clean waste oil

http://pabiodieselsupply.com/shop/oil-cleaning-centrifuge-motor-package/

This company makes a unit capable of cleaning 300 GPH. Click on the waste oil link, and scoll way down to the bottom. Apparently, the cleaned oil is adequate for powering IC diesel engines.........

So, my question is, would running our waste oil through a cleaning process result in something clean enough to burn in a vaporizing rather than atomizing burner?

Dave

I'm sure I've read that one of the international tourist railways does this- Ffestiniog, Puffing Billy, or West Coast Wilderness Railway? I'll bet Nigel will know; hopefully he'll check back in before too long.

Hugh

Thanks, I haven't had much luck tracking this down........

The more I think about it, even if not clean enough for vaporizing it might make for at least a more relaible and higher quality fuel. If the centrifuge were mounted in the tender tank itself, it would be a simple matter of hooking it up with an extension cord, and letting it run for a couple 2 or 3 hour intervals during a workday, until there's nothing much in the way of grunge left in the trap. Not a time and effort hog......

Dave

I've been scouring the internet to see what I could find this morning. The Ffestiniog apparently has switched back to coal firing for most, if not all of their locomotives, but they did use waste oil for a considerable period. I found several references (but no details) to their practice of centrifuging the waste oil to purify it.

I found a forum with a good discussion on using waste oil as fuel for diesel engines here: http://www.survivalistboards.com/showthread.php?t=212731

There are lots of links in the thread; I'm running through those now to see if there's any directly relevant information. Here's one on filtering waste cooking oil, but it has a lot of good info most of which is applicable to filtering any waste oil: http://www.dudadiesel.com/filtering.php

Hugh

I visited Ffestiniog several years ago - shortly after their oil conversions and was of course interested in how they did it. That's when I saw my first vaporizing central burner arrangement, which was very impressive in terms of its cleanliness and efficiency. The little 0-4-0T could haul 2 additional cars on oil that it couldn't do on coal, and made minimal smoke. For some reason it also could power a Nathan 5 chime whistle that was about half the size of the boiler........

In any case, back then I was corresponding woth a young mechanic there called David who showed me around, including two burners he designed built experimentally. We also worked with another crusty gray bearded guy who could have been my brother hanging rods on the Garrat. Good times, good people interesting work being done.

What i don't recall is the use of cleaned waste oil. Perhaps that came along later, or I just forgot to ask.

I think if we could make contact with somebody working there, we could find out about whether centrifugally cleaned waste oil worked in their vaporizing systems, and how they set up to do it. Do any of our contributors have any contacts there to ask?

Dave

Hi Dave,
I have been in touch with FR right after my book, 2006/2007. I'll try and revive the contact. In the mean time, their website gives a contact address for the Boston Lodge Works where they should know more:
Email: tony.williams@festrail.co.uk
Phone: 01766 512850.
Kind regards
Jos

Thanks Dr K. I'll try an email to Boston Lodge and see if I get an informative response. Generally a contact person makes for a better response, though, I appreciate your kind assistance.

Dave

First answer received! It appears that initially normal oil was used, after that waste oil was gathered by volunteers but they had to buy processed waste oil from outside sources. All loco's but two are changed back to coal given the oil price.
Kind regards
Jos

OK....by "normal" I'm thinking of a lighter grade of fuel oil. Not sure if the need to purchase waste oil was due to insiufficient amounts available for gathering or trouble getting it clean. Good to know two are still oil fired, so current knowledge should be available on the burner and drafting systems involved and eactly what processing is done with what result.

I emailed in an inquiry - hope details wil be forthcoming.

Dave

One obvious thing on filtering the waste oil (if it's not already being done); the water will separate naturally and collect on the bottom of the tank, given sufficient time. If the tank is located above ground and there's a drain valve on it, it's easy enough to drain off the water occasionally. Apparently heat and/or the addition of a thinning agent (such as a small amount of gasoline) will significantly speed up the process. Since the railway is basically a one weekend per month operation (correct?), I'd think this practice would work pretty well.

Hugh

DaveLathrop57 wrote:..... Good to know two are still oil fired, so current knowledge should be available on the burner and drafting systems involved and exactly what processing is done with what result.....
Dave

Dave, don't reckon yourself rich! This was the comment:
"The conversion to coal was a huge success and has now been adopted as the
policy for both railways. All the WHR locos have been converted.
Most FR locos have been or are in the process of conversion.
There remain some problems with Linda and Blanche which have very small
grates and were the reason for conversion to oil in the first place.
So they are still on oil but hardly used because oil is now too expensive!"
Kind regards
Jos

I suppose it would be needlessly simple to go back to GPCS combustion and Lempor exhausts as Linda was previously equipped!

Hugh

Oh, well....if they are kind enough to assist us, there should be drawings or other documentation as well as the memories of the firemen and mechanics.

Interesting point, Hugh - was the Lempor removed because it was found somehow incompatable with the oil burning?

I think centrifugation of the waste oil is a good idea, if the money can be provided for a rig at appropriate scale.

Here is a discussion of some of the chemical effects found in the usual kind of waste oil (scroll down for the relevant section):

http://www.machinerylubrication.com/Read/193/oil-glycol

Inherent in the discussion is that water is added to the mixture, and perhaps 'emulsified' with it, to remove some of the chemical contaminants more effectively when the mixture is centrifuged.

I haven't seen reference to this yet -- but there is a video library of information on centrifugation at this perhaps appropriately-named location:

http://www.centrifugefordummies.com/

This promises to tell all about how to construct the equipment -- they indicate that it is possible to process waste oil to make it suitable for diesel fuel (no particulates above 2 micron size) which is more quality than required for modern types of high-pressure/spray burners.


Finally, there is a library of very interesting-looking links at:

http://www.pabiodieselsupply.com/link.htm

(scroll down to the section on waste oil filtering).  Anyone wanting to get a quick education in the details of waste oil as a fuel should not have to look far beyond these references...

I don't see any reason not to include a centrifuge simply because it greatly reduces the potential problems with a very variable quality of fuel. We have no fuel storage tank at the railroad; all the oil we collect goes right into the tender. 110 has no fuel tank left in the tender, so we need to build a new one. Including a centrifuge system in the new fuel tank design would work very nicely I think. We have a larger manhole with the centrifuge in the opening, and plumb a pump and line to it from the sump. During workdays, the centrifuge can be run for a couple hours on its own, cleaned out, then again, 2 or 3 times until the fuel is clean. It won't eat up a lot of crew time for operational benefit.

What I would do in this particular circumstance is have multiple centrifuges mounted on the manhole cover -- you can probably get seven or more on there if you wanted -- and just leave them there.  (They have mounting-bolt holes on the bottom, and are attached directly over whatever it is that is used as 'tankage', so there is no added cost or difficulty, and the only real design 'constraint' on where you mount them is having enough material to keep them steady in operation -- note that the PA Biodiesel units appear to be using the head of a 55-gallon drum quite happily as a 'mounting plate')

The reason for having multiple centrifuges is to cut down on the 'dwell time' of whatever was used to deliver the waste oil.  I would strongly advise the purchase of a couple of those portable poly tanks (I know where several discarded ones, formerly used for farm service, are just sitting abandoned, and I doubt the cost of the things are excessive to begin with) and perhaps a cheap prefab storage shed to keep them neatly in, and use these for the oil supply.  You can see an illustration of one of these tanks on the PA Biodiesel page that references the tank heaters.

There really is little reason not to filter the oil from one of these tanks into another, and put the centrifuges and pump(s) right in the shed, under cover and neat.  Yes, it's more infrastructure, but note that there are other uses for the centrifuged oil than just burning in the locomotive. Quite a bit of diesel-engined equipment will run happily on the centrifuged product...

One additional note:  You may want to perform staged separation if using added water to ensure glycol-byproduct separation and so forth.  I see this involving a first 'dewatering' stage, with the oil supernatant then going to the oil centrifuge system under pressure.

Oh, by the way, we just acquired yet another compelling reason to restore the tender first, and to restore the tender quickly ... it can serve as part of the oil storage for the other locomotive while the rest of the 110 project is proceeding...

There are legal environmental issues with stationary oil storage tanks we'd like to avoid.......but a tank in a tender isn't stationary, so that would work well. I'm thinking of one of the 300 GPH centrifuges here, so if we build a 600 gallon oil tank it can run all of it through in 2 hours.

I need to learn more about filtration - seems if we settle and then centrifuge first, the majority of the nasties will be gone before we filter, thus maximizing the life of the filters.

Dave

>I need to learn more about filtration - seems if we settle
>and then centrifuge first, the majority of the nasties
>will be gone before we filter, thus maximizing the life of the filters.

It was my understanding that the centrifugation WAS the filtration, to the extent required, and that no post-filtering of any kind would be involved.  The exception would be to use some sort of coarse (cloth?) pre-filter in the line going into the centrifuge unit, but I suspect the strainer or 'sock' on the pump pick-up, and the characteristics of the transfer pump itself (which must be a pressure device to develop the amount of 'spin' to make the centrifuge work) would determine small enough 'particulates' before any inline filter would be needed.  I'd just make up a big prefilter box or sock with the desired aperture size and do the prefiltering with that.

Remember that there are only two reasons we're centrifuging:  to separate any liquid or entrained contaminants (including with water or polar-solvent flush), and to remove some level of particulates or solid contaminants.  The amount of particulate reduction given in some of the literature indicates centrifugation works down to 2 microns.  This is diesel-injector level, and I believe even a sophisticated rotary burner can tolerate particulates much larger than that.

What is not indicated in the material I have seen so far is how many centrifuge passes would be required to get the oil down to a 2-micron level.  One 'theoretically' easy way to do this is with multiple sequential centrifuges... but you'd need a separate pressure pump between each stage.  There isn't any reason why multiple pumps could not be driven off a common motor, though.

I certainly see very little reason for going to any kind of spin-on or element filter in the fueling system!  The one thing I wonder is whether some design of centrifugal separator could be made that would continuously 'dump' the separated contaminants, either fully unattended or semi-automatically.

My limited research shows cents have housings that collect the grunge and must be occasionally cleaned, and that doing that occasionally until little is found is the usual small batch process - small being up to 300GPH in the sources I've located. Running sequential cents or cents in parallel - parallel isjust making a big one out of little ones, and series isn't easy to make since the cented oil dribbles out of the housing back into the reservoir. A stairway system would be too vertical to fit on the top of the tank.

There may be other hardware out there I haven't found yet.

Dave

Series involves gravity drain -- nothing more complex than a P-trap arrangement with the next pumping stage at the lowest point so it's self-'priming'.  There is nothing that says the output from a centrifuge stage (I would use 'fuge for the short term, as 'cent' is already taken for coins and hundredths) has to drain directly into the reservoir, and of course each subsequent stage will involve lower volume, so there is little chance of oil backing up to an earlier-stage rotor lip and stopping the separation.  If you were mounting the centrifuges on the tender, all but the last one will be bracketed up above pump level; the last one of course ports directly to the reservoir.

For additional strings, you run in parallel -- there are several pictures of this in the complete-rig section of the PA Biodiesel site, for example.  They apparently split the pressure flow between two centrifuges to get the higher throughput; I just extrapolated to multiples in case higher short-term throughput (as, for example, might be the case if you had to process oil from a typical home-oil-size tank truck in 'realtime' without causing long delay) needed to be achieved.

I think I'd use the inverted-rotor, heavy-case version of the centrifuge, unless you need to worry about impact damage.  It might be sensible to build an outside 'case' out of, say, plate sections, or a safety cage, to protect a cast-light-alloy centrifuge case, and associated piping,   from the usual sorts of accident (including dropping the tank 'cover' to the ground by accident!)  If series stages are going to be used, the pump(s) ought to go somewhere on the tank as well, because otherwise there will be lots of fairly long lines.

I have been ASSuming that the same general principle used in bypass filtering of diesel lube oil is applicable to centrifugation (see the first reference I posted above).  For bypass this says as a rule of thumb that about 7x the desired volume has to be cycled through the filtering system and 'back into the sump' to get low concentrations of what is being removed.  But that is with a filtering medium that achieves 100% filtration in one pass through the filter.  If centrifuges take more passes to accomplish the business -- and I see some indications on the Web site that they will -- then you'd have to do some further math to figure out how many times the volume would have to be circulated to get desired fineness if the first stage or pass goes directly into the tank.

That does not mean I think less of the centrifugal system than a good bypass filter, just that we need to get the scaling factors right.