well, I have been rumaging around, collecting pieces for this project, which i believe is a great invention,tim.
first off i have an actual hyd. pump i picked up at a sale for $25, and i found some oriface' which came from a stove, i beleive. they are marked 65,68,70. while these may be too big, since i have them i will try them out.
I have a 4500 watt element left over from other greasey stuff, as wel as lots of pipe.
i am only lacking the temp controller. what is everyone having luck with. i remember reading about watlow's, are they still the best? i would like to find something cheap. the town i grew up in has a watlow plant- maybe time to call some old friends and see if anyone works there...
some seem to be using a furnace pump- too bad we cant derive a temp reg from a boiler or furnace, get all your parts from the same source.
Be the change you hope to find in this world.-Gandhi
Wow you got 15 pages and no mention of this:
Tests at this point say a mistwash is required before this rig (or any evaporative method) will dewater any VO that contains large amounts of suspended acids/salts/sugars/etc.
That was brought up by Dana on page 1 of the Dieselcraft post and almost every other page of the thread too.
Also that the pan test is not reliable after evaporative dewatering. Since this method is very similar to the Dieselcraft, I was wondering if you all are using another method to test water content besides pan test after the FE?
Have you run any lab tests or other tests to tell whether any suspended stuff is removed or just concentrated to where the pan test quits working? If you have answered any of those questions it might apply to other evaporative methods too. Or at least how you did any tests would be useful since we are doing tests and lots of discussion like that on that thread, and this seems very similar and we could help each other test these questions.
I didn't read back through all of this discussion but I think Dana surely made the same comments someplace along the line. I think I also mentioned that I do water-washing to wash out any water solluable stuff prior to using the FE someplace along the line ?
I have not had any lab-type tests done for water content, I do a 10% water wash, with one pound of baking soda dissolved in the water, as my first cleaning step so I am ASSUMiNG this removes the bulk of any sugar, salt, water soluable acids etc but no lab testing to confirm it, It has to be better than my prior unwashed oil. I then filter the warm/ washed/settled oil very well (25-30 passes through 5 micron filters), then run it through the FE.
My only test for water content is the hot frypan bubble test, it was originally toughted to be an acceptable water content test, there is an extensive discussion someplace on the forum, mainly between Dana and Neutral, as to whether veg oil contains free water, microdroplets, or dissolved water, no real conclusion there but the frypan bubble test (not simple "crackle" test) seemed to be acceptable to indicate any types of the verious water content types.
The "concentrated acid" causing bogus results with the hot pan bubble test issue is a new issue that I think has just recently been raised for the first time in the dieselcraft centrifuge thread, seems like it may be true but I have not yet seen any detailed discussion of this. Since I water-wash prior to the FE the "concentration" situation is something I has not been concerned about.
Could some of you that built them , please post pictures of your variations of FE set-ups . - Thanks
1994 F250 IDI 7.3 NA E4OD
Remember that the forum search/"find" feature does not include the archives . Search the forum archives here-
I agree that the wash probably makes it OK.
I think the concern is concentration of anything thats in there like sugars, salts, etc. not just acid. And so I think Dana's issue was for someone who didn't know that the wash was "required" (if it were not stated that way on the first post and most every page following) would use an evaporative rig without wash, and also not know that the pan test can fail to work properly in that case.
I thought that heated settling for >24 hours with no wash was sufficient for most people to remove suspended water? Have you tried doing no wash, and just heated settling on some of the same VO you are processing through your FE as sort of a control test? (since you are doing lots of heated settling anyhow.)
My rig has lots of pics in the Dieselcraft thread you can see, and uses a FE similar to this, with higher flow rates. And check pages 6 and 7 for pics of another rig made by JOAT and also closeups of the FE parts of mine.
Many thanks Tim for all the work on this and posting it here you can see it was big part of what inspired me.
For the first several years I only settled my oil at ambient temps then filtered it several times to 5 micron, I recently ran across a sealed 5-gallon jerry can of this ambient-settled-only oil on the bottom of my can stack, It had no visable water showing, I ran it through the FE and got more steam than I was expecting, about the same as with my current warm-settled oil.
I also have been running the contents of a sealed 400 gallon tank through the FE. This is oil that was only ambient-settled and filtered, at least 3 years in storage. There was absolutely NO free water present on the bottom of the tank. I was plesantly surprised to find that even the very bottom-most oil was no wetter than my current water-washed/warm-settled oil.
I get steam from every batch of oil when running it through the FE, more heat produces what appears to be more steam but I am suspecting a lot of this is actually oil vapor when you get up around 350 deg. I am finding a slight oil film on any cool metal that is close to the steam outlet even with my steam output oil-condensing tank setup so it is possible to still pass a bit of oil vapor out to the atmosphere at this temp.
It is interesting to read that you get less steam with each successive pass of oil through the dieselocraft centrifuge and the steam is gone by the 4th pass, this with 200 deg or less oil, I may be heating my oil way more than nescicary at 300 deg ? Have not taken the time to do the required testing, easier to just run at 300 deg and be confident I am removing all the water, may not be the most efficient power-wise but easier (Well, maybe,having filter problems, read on) .
I may take the time to test the lower temps as I am finding that at the 300 deg temps I appear to be doing something to the oil that clogs the cintered bronze filter. To insure clean oil to the FE I am feeding 5 micron oil to the FE directly after the last 5 micron filter on my filter barrel setup, the cintered brass filter on the FE is rated at 90 microns, it should not be plugging but it is. The cintered brass element became so clogged that I have been unable to clean it by any method so far, I removed it entierly and then began clogging the .018 orifice after about 4 hours of running, happened several times while running the 400 gallon tank of oil through the unit -SO- it looks like 300 degrees causes polynerization on some portion of the oil, if it was a simple thickening of the oil I should have been able to unclog the filter using warm solvent while backflushing it but this did not help. whatever is clogging the filter has been altered such that standard solvents doesn't thin it, I assume this to be polymerization or vulcinization etc ?? whatever it is it ruins a costly filter element by plugging it up entierly.
Lower temps may eliminate this problem but at the expense of requiering multiple passes through the unit. Running at a higher flowrate with a larger orifice diamiter and no filter would likely work without plugging the orifice but it also requires more heat, this added heat may cause more polymerization ?
I have backflushed the heater pipe several times since feeding it known 5 micron oil for several hours, each time I get about a teaspoon of small bits of stuff from the pipe. This polymerized oil is about like sawdust for size, no big clumps but lots of tiny-but-visable translucent bits. It looks like the final dewatered oil should be filtered one more time to remove the tiny bits of polymerized oil.
Obviously still a work-in-progress.
Was there any suspended water in that? Did you try a pan test before FE?
It would be interesting to try some heat settled, no wash of the same source just to compare the pan tests before FE. Since it seems the majority of people are doing the heat,settle,filter method. Maybe you could get good results with just the heat settle and avoid the baking soda and all the extra time that takes.
And even more surprising to me was that a large amount of steam puffs out even at 120F, you might want to experiment some with lower temps. Although it is likely that the centrifuge is combining small suspended water particles and building up in the rotor and then spits it out as a larger spurt, which is how it sounds, its real intermittent and when the steam comes it spurts a large amount. Which makes me think its not just tiny droplets evaporating 1 by 1. On mine it didn't reach the 200F until the last pass, so most of the dewatering was occurring much below that.
Suspended water -- (or dissolved water or microdroplets ??) I did not do a pan test on the old settled-only oil prior to running it through the FE but I am confident there would have been some amount of tiny bubbles produced in the pan, every other test of settled-only oil that I have done has always produced some amount of tiny bubbles. The only oil so far that produces absolutely NO bubbles in the hot pan test is oil that has been through the FE.
I suspect that very tiny droplets of water(smaller than are visable to the naked eye) just dont have enough weight to gravity settle out of even hot oil due to buyance, surface tension,viscosity, and temp differance currents etc.
The dabate as to how harmefull these tiny droplets are has yet to be settled, technically they can cause cavitation erosion in the injection pump and injectors or anyplace where there is fast changes in pressure, but as to how this lessens the life of the parts in real life is going to be pretty hard to define. Past discussions about this are here and here, there are more, search "microdroplets".
How close to the heating element are you measuring the 300F? You are probably getting much hotter than that at the surface of the hot element, which could be causing polymerization. Does your heating element get any black build up? That would indicate polymerization, but it could occur without build ups too.
I am running 1gpm flowrate at lower temps so have seen none of that. I am running I think the same 4500w/220v elements at 110v = 1125w but 2 of them.
Higher flow rate, lower temps, more passes, sounds like you are getting closer to making a Dieselcraft rig
I recently changed my heater pipe to a vertical pipe mounted next to the flash tank, the temp sensor is at the top end of this pipe within a couple inches of the end of the heater element. I don't see any buildup on the element but the presence of the tiny polymerized bits does indicate that the surface of the heater is hotter than the general oil flow across the sensor, sorta have to expect that since oil has only about half the heat transfer rate as water. Big differance with your dieselcraft setup is the much higher flow rate across the heating element, the oil does not stay in fixed contact with the surface of the element nearly as long as in my 5 G/H setup, you are moving oil 10 times faster than I am, may be the way we have to go to reduce clogging ?, sorta looks that way so far.
I gave up on dewatering the entire 400 gallon tank due to clogging, I will lower my temps down to 150 deg f or so and try again, if this eliminates the clogging and the poiymerized crud but still removes the water it will be a good indication of future tests.
Lowering the tempmay not actually reduce the polymerization with my current temp controller, the heater element is still turned full-on or full-off at some fairly slow rate ao several seconds so the element may still have time to overheat the oil that is in direct contact with it's surface. A true proportional control would reduce the surface temp of the heating element but this requires that the size of the heating element be more precicely balanced to the flow rate, simply increasing the flow rate of the oil would be the easiest way to eliminate this problem, surely the cheapest.
Agreed and I think much better since you don't have the element cycling on and off which can't be good for consistant results. Plus you get more done quicker. I think adding more orifices would be better than 1 larger one as a way to increase flow.
Thanks for all the good work here, Tim and all. I haven't seen this question asked or discussed, if it has, my apologies in advance.
But...rather than heat/pressurize the oil and all the related problems, what if you heated the bash plate to 275 degrees F and bashed a stream of 140 degree oil against it? Wouldn't the water flash off and the oil run down the bash plate? Or even discharge from the Dieselcraft onto a heated bash plate? (overkill?)
2002 F250 Vegistroke now with the new V3 module!
Heated bash plate dewatering -- This might work but would be based more on the "hot surface causing evaporation" principal similar to using an electric skillet than on the "superheat the water under pressure so it flashes off at normal pressure" principal. The problem I see is that the oil would be in contact with the bash plate for such a short time the plate would have to be realy hot to cause the water to evaporate, this is likely to agrivate the oil polimerization problem that I have been agrivated by from the start of this project.
High heat -- I have been having considerable agrivation and problems due the the 300-350 deg f temps that I have been heating the oil, it causes enough oil to polimerize, or vulcanize, or at least thicken, enough that I have clogged 2 of the cintered brass filter inserts thus far, the first insert was a 10 micron unit, it plugged beyond cleaning by the end of last summer, only a couple months of use. I replaced it with a 90 micron element and got through this summer but it became clear that it also was plugging even though I was solvent-cleaning the warm element prior to each time of shutting down the FE.
These filters are plugged with something that does not dissolve out, even when using several differant warm/hot solvents and pressure backflushing for hours, it is more than just thick oil, something has caused the oil to change enough that it is no longer dissolvable, I suspect the oil has vulcanized into a crude form of natural rubber ?
I have been deliberatly feeding 5 micron filtered oil directly from my filtering setup to the FE most of this summer to insure I was not clogging the FE filter with particulate, this convinced me that some small portion of the hot oil is somehow being physically changed.
I have also plugged the hot oil gravity-flow output hose of the unit a couple of times, both times the hose was plugged with brown pea-sized rubberlike chunks of what I suspect to be polimerized oil.
I previously reported here of finding one large single thin index-finger sized chunk of similar stuff in the bottom of several of my storage jerrycans after the dried oil had cooled. Whatever it is it has a very high attraction for itself as I only find larger clumps of the stuff rather than small bits ?
Reading the "Dieselcraft centrifuge for filtering/dewatering" thread (here) they appear to be having success with flash dewatering at a consideably lower temp. It is indicated that they see steam at as low as 120 deg f. They are passing there oil through the unit several times and indicate that steam decreases with every pass and the oil will pass the "almost no tiny bubbles" in the hot frypan test by the 3-4 pass. I am trying this technique with the FE, the temp of the oil coming from the heater pipe is now set to slightly above boiling (215-220 F), the temp of the hot oil exiting the flash tank has measured around 150 deg f for the entire 24 hours that I have circulated this barrel of oil.
Initialy in daylight I was not seeing any steam (was looking for the normal-with-the-high-temps large amount of steam) but when I checked after dark with a flashlight it was apparent that a noticable amount of steam was flowing from the vapor outlet pipe and the oil vapor condensing output was dripping one drop every several seconds so both water and some amount of oil were being vaporized even at these lower temps. Watching closely in daylight today I was also seeing steam, just not nearly the amount that was being produced at the higher input temps. I am recirculating the oil several times in this low-temp testing and will do hot-pan test both before and after each pass in the future (several barrels still need dewatering before it gets realy cold here).
Hopefully this works ok as it takes less electricity to make the lower temps and will hopefully eliminate the plugging of anything with small passageways due to the thickening/polimerization/vulcanization of some small amount of the very hot oil that I have been working with up til now ?
Tim and others - very nice work and postings.
Would you mind listing the settings for your PID temperature controller ? It MIGHT be possible for us to calculate an improved setpoint, depending on how your controllers actually are operating. Some are true PID analog control, but many are actually pseudo PID, and the I is replaced with a sampling rate (timing). This might be the case since your output is driving full on / full off.
There might be a way to set the controller to cycle the elements on / off at a higher frequency, and "clip" out the on cycles that are not needed.
HarryN -- I am not using an actual PID temp controller, just a simple electronic on/off control with about 8 degree sensor temp deveation including the hot/cool overshoot. I might try using a simple-but-brute-force variac to lower the AC voltage to the heating element but the element temp will still likely be too high. I think the best thing would be to use a heating element with a much larger surface area to reduce the actual contact temp but this is much more difficult than using a simple water heater element. Just another thing to experiment with.
It is actually looking like the Dieselcraft or spinner II pump-pressure driven style centrifuge approach is the more practical way to procede with future dewatering development. The initial cost is considerably more but it allows most folks to easily build a working device themselves without all the custom fabrication required for the flash evaporator, and it looks like the maintenance is going to be a LOT less cost and hassle.
The centrifuge works to filter out the bits prior to the oriface without requiering the costly replacement of a static filter, and the lower operating temp used to flash off the water is low enough to not cause additional polimerization of the oil. It does require a few passes of the oil for dewatering rather than a single pass but my testing indicates the only way to get dry oil in a single pass is to push the heat up around the 300 degree f mark, unfortunatly this temp causes the polimerization that causes the inside surface of the plumbing to become coated with a "skin" of rubbery oil that brakes away from time to time and the bits clog the oriface and the in-line filter, high-temp filters are expensive.
If one-pass dewatering is required it looks like it is a given that it will take a good bit of maintanence of the flash unit and the dry oil will also need to be filtered again after passing through the unit. This sort of explains why I have been finding a flat-but-finger-sized clump of plastic-like stuff in many of my storage jerry-cans, and have plugged vehicle filters regularly, even though the oil had been well-filtered to 5 micron prior to being run through the flash unit.
I recently replaced my vehicle filter due to plugging, looking closely inside I found considerable visable semi-transparent bits of polymerized oil so there likely was a lot of the same tiny stuff clogging the actual filter element.
Live-and-learn, that is part of developing a new product or approach to about anything, some work better than others. I will keep testing with this unit, next testing will be at a lower temp and multiple passes to see if it reduces the polimerization problems.
Spending time now trying to get a vegoil fired oil burner up and running for winter. The oiltemp/polimerization point info will be useful in building and testing the heater-in-a-pipe oil preheater for this project.
What is the current thought between the Diesel-Craft and the Spinner II. I read earlier that it was the cost. I am currently checking out a couple of different units. I am leaning toward the spinner II. I am going to get something in the 200 gph range. the dieselcraft at 108 gph is 383.00 300 gph is 1560. thats a bit too much for me. (the 300). I have not received prices on the spinner yet
anyway, whats the thoughts
Elvia the HandyMan
2000 f350 XL utility/service truck
There is only 1 person (fattywagonman) who has posted comparing the spinner to the dieselcraft, and that is on page 14 of the dieselcraft thread here. The small Dieselcraft I am using at $230 was less than half the price of the small Spinner at $495.
The dieselcraft centrifuge discussion (here) has been having good luck with low-temp water evaporation so I have been trying this with this FE (also because of all the plugging problems associated with the 300 deg f approach).
I changed the way I am using the FE unit now. Originaly I first filtered the oil then ran it through the high-temp FE, this was causing problems due to the high temps polimerizing some of the oil and clogging filters, orifice, plumbing etc and also leaving bits of plasticized oil mixed in with the final dry oil - All not good things.
I have now been running the FE at much lower temps and in paralell with filtering. my filtering setup circulates a barrel of oil through a bank of filters continuously for at least 48 hours, usually 72 (or whenever I get back to shut it off). I now pull oil for the FE just after the last 5 micron filter in the bank, some of this oil goes to the FE, some goes back to the barrel directly. This way I am assured that the FE is getting clean oil. I have removed the cintered brass high-temp filter completely from the FE, and have slowly been lowering the temp while inceasing the FE's pump pressure (This approach was discussed earlier in this discussion, trading pressure for temp, it seems to work). I have not had any plugging problems of any kind since lowering the temp below about 250 deg f and still get a lot of both water and oil vapor released.
I now have the pump turned up max to 170 pounds and the temp going into the orifice is set to 160 deg f, the temp coming out of the orifice is about 125 deg f. There is still vapor being released, not the fogs of it that are seen with the 300 deg f temps, but enough you can see it with your eyes and especially if you wait til dark and shine a flashlight at the vapor outlet. The oil vapor condensor is also dripping a lot of condensed oil so vaporization is definitly taking place at these lower temps.
I now run the FE for as long as I filter the oil, the oil ends up clean and dry without the extra step of using the FE seperatly and the plugging and polimerized oil bits problem has been eliminated.
On the next barrel of oil I hope to change the spring in the pressure regulator in the FE oil pump so that I can get the pressure up to at least 200 pounds, preferably 250 pounds, this should let the temp be reduced even more, down into the range where the oil can be heated entierly by radiator water for on-vehicle on-the-fly dewatering, I HOPE..
Would you like a copy of a temp controller circuit?
I got a copy off Matt and he is happy for them to be passed around.
mathematical elegance -- desired result achieved with minimal complication