I'd put the thermocouple on top of the barrel because heat rises. Make sure you insulate the barrel top ans sides. Most of the heat is lost thru the top. While a PID /SSR thermocouple is nice, I think it is overkill for a heat/ settle system. I'd use a hot water heater element thermostat in series with the heat trace cable. That's good enuf.

quote:

Originally posted by DCS:
Just attach the end of the thermocouple to the side of the tank with some metal type putty or similar. All it has to do is transfer the temp of the tank to the thermocouple.

Evenn if it's out a degree or 2, it should be constant and you can adjust your controller to compensate.
It's not like you need a reading accurate to 1/10th of a degree.

I agree with DCS. Stick it on the side between the turns of the heat cable. About 2/3 of the way up is a good spot.

DO NOT place the TC on the top. An air pocket could effect the reading and the air pocket could change. Accuracy is not really important but consistancy is.

Contrary to common belief, heat does not rise. Heated convection flow rises but heat itself doesn't. For settling to be effective temperature needs to be very even to prevent convection flow.

A mechanical thermostat will work but it needs a small differential to keep temperature swings and the resultant convection flow to a minimum.

It's usually best to parallel upflow tanks rather than series cascad them. You want to keep the flow rate as low as possable. Cascading past 2 has shown little effect. I have had good success with one heated upflow settling tank capacity equal to 5 days of use.

quote:

Cascading past 2 has shown little effect.

What little effect?

Galt, You think 4-275 totes will allow enough volume for processing over a 100 gallons entering the system at a time? I would be using like 6" sewer pipe for all of the down tubes.

Anyone ever experiment with a pump at the beginning of the system? That way one can load the raw wvo container and with the pump know the flow rate. Just an idea but figured someone's tried it.

quote:

Originally posted by Ron Schroeder:
It's usually best to parallel upflow tanks rather than series cascad them.

I have wondered if there was any difference being that the amount passed through would be equal either way to whatever was put through.
I had guessed that putting them in parallel was better because it split the flow and made for longer residence time in each tank and less possibility of stirring.

That said, my setup is in series through available space considerations. I was going to re-arrange my setup but due to the fact I have about 18 months capacity at my present usage rate and only flow the oil once a month, it's pretty inconsequential for me.

I have been adding oil to the top of the second tank from the first and believe this may have benefits. As my first tank was outside in the sun and reached a higher temp then the second tank always shaded in a shed, I believe I was getting currents from the warmer oil stirring the second tank up. By adding the warmer oil to the top of the tank, I believe it just allows anything in the added oil to drop out without bringing any contaminants from the bottom.

I draw off what I need from the top of the second tank, usually about 1/4 of its capacity, and then top it up again where it remains undisturbed for a month till I need more oil.

I did try an experiment at one stage sequencing 5 25L drums but also found little difference between the oil in the second and the 5th drum.
My experience leads me to believe that residence time is the most important thing over all. I also experimented with tall and squat tanks and have come to believe that a taller tank in a relatively constant flow system may in fact be better than a squat tank of even greater capacity.
It seems that the introduced oil, no matter how slow it is fed in will stay in a relatively small "lump" and spread vertically easier than spreading evenly around the tanks diameter if it is a squat tank such as an IBC or " tote".

I am of the thought that even though a squat tank of more than double the capacity should have theoretically less vertical movement, a taller tank with greater separation between the introduced oil and the output will allow faster throughput of oil to the same or even better cleanliness and dryness. Cascading shallower tanks would seem to give the same effect of depth but I have resevations the theory works the same in actual practice.

I also experimented with placing a tall tank outside in the sun and insulating the bottom 2/3 and painting the top 1/3 rd black. My aim was to thin the oil at the top at the same time as limiting any convection currents at the bottom of the tank. My thought was that the thinner oil would allow contaminants to drop out more easily into the cooler oil lower in the tank without them being brought back up again even if there were a swirl motion at the top in the mainly clean oil.

My observations were not definite but did give rise to thinking that the theory may have some merit particularly in tanks with good height to width ratios. This was borne out by an increased throughput of oil than what I was previously getting that was still very clean and just of the same moisture level.

quote:

My experience leads me to believe that residence time is the most important thing over all. I also experimented with tall and squat tanks and have come to believe that a taller tank in a relatively constant flow system may in fact be better than a squat tank of even greater capacity.

My experiments show similar conclusions.

quote:

upflow settling tank capacity equal to 5 days of use.

..or amount introduced per day, is a good rule of thumb for sizing an upflow system.

Of course a lot depends on the quality of the oil you get. Unless your source is draining the used fryer oil back into cubies you can expect that almost anything has been dumped in the 'garbage' along with the oil. In the vast majority of 'food service businesses' staffed by minimum wage idiots, there is a dumpster for solid waste and a barrel or dumpster for liquid waste like "WVO" and any other liquid garbage.
If you're sucking your "WhateverVegetableOil" from a barrel or dumpster then mist washing and settling before any other processing is a very good idea.

Okay, well I'm looking at an upflow system purely as a prefilter before centrifuging and final filtering. I will be processing about 2000 gallons a month so I will be dumping at over 100 gallons at a time. I am hoping to remove just most of the particulate and water but can take care of what remains. Any idea on sizes for down tubes and top tubes?

quote:

Any idea on sizes for down tubes and top tubes?

1" will be adequate. Pipe size isn't important, flow rate is. 2000 gal per month with 100 gal per fill is 100 gal every 1.5 days, so you want to adjust the flow so it takes 1.5 days for the 100 gal to flow through the upflow cell. Approximately 3 GPH.

quote:

Originally posted by john galt:

quote:

Cascading past 2 has shown little effect.

What little effect?

Carl Fischer testing for water after the first, second, 3rd, 4th tank etc. has shown that very little extra water is removed after the second tank. If you are adding more tanks in series to handle extra flow, the low limit of water is not as low as with lower flow and adding stages doesn't help much. Paralleling tanks so each one has lower flow gives lower PPM water readings for the same total volume of settling tanks.

I have found very little differance between 1:1 height:diameter tanks and about 2.5:1 tall tanks but have found worse results with both very tall and very squat tanks. It's not just residence time, with tall tanks (actually any tanks), the upward flow MUST be slower than the falling speed of the smallest particles that you are trying to remove.

For short tanks, I have found diagonal flow to be best.

I usually use a drop tube of 1/2" pipe but neck it up to a larger diameter about 6" before discharge to reduce turbulence.

In general, I have found direct solar heat to cause too much convection flow for good consistant results in one tank. Sometimes it works, sometimes it doesn't. Even heat on all sides of the tank is important.

I do run my inflow thru a tube in tube heat exchanger to get the oil to within a few degrees of the settling tank (slightly cooler is better than slightly warmer) to prevent convection currents from temperature differential.

quote:

Originally posted by dlor1:
Galt, You think 4-275 totes will allow enough volume for processing over a 100 gallons entering the system at a time? I would be using like 6" sewer pipe for all of the down tubes.

Anyone ever experiment with a pump at the beginning of the system? That way one can load the raw wvo container and with the pump know the flow rate. Just an idea but figured someone's tried it.

With even enough heating and the proper temperature, you probably wouldn't even need the CF with 4 totes at 2000 gallons a month.

Do you also want to remove the fat or are you going to burn the fat?

quote:

Originally posted by Ron Schroeder: It's not just residence time, with tall tanks (actually any tanks), the upward flow MUST be slower than the falling speed of the smallest particles that you are trying to remove.

To me that sounds very slow. It makes sense but to picture it in a totally constant flow system brings up mental images of a flow rate equivalent of something gravity trickling out of a drinking straw. I have always found the settling times generally quoted to be much faster than would give me the results I want.
What people quote in multiples of weeks, I often equate to months and I don't think the oil is so bad as to be responsible alone.
Perhaps I'm just too fussy.

quote:

I do run my inflow thru a tube in tube heat exchanger to get the oil to within a few degrees of the settling tank (slightly cooler is better than slightly warmer) to prevent convection currents from temperature differential.

This is a good Idea. The thought that comes to mind is for me to install a holding tank next to my upflow tank to allow the oil to cool to the same temp before adding it. Next thought is more complication, more space needed, more fittings, valves and pumps, more Aaaagggghhh! factor.

What I have done is to start putting some lengths of roofing tin on the top and sides of the cages of my IBC's to at least keep them shaded from any direct sun exposure. This should be effective in bringing the temp of the tank in the lean to, much closer to the temps of the tanks outside.
Hopefully!

The sun exposure is limited to a few hours a day so hopefully the air gap between the tin and the tank itself will dull the heat transfer sufficiently to alleviate the problem, especially in winter.

Hi DCS,

Settling time is very dependent on viscosity and therefore temperature. The flow is just a trickle but it is for 24/7 by gravity so the daily throughput is still reasonable. I do over 50 gallon every week with a 60 gallon settling tank. I have run thru over 30 gallons in two days but have not tried over that rate continuously.