Homemade centrifuge

[patracy]

Just a thought, the turbine and impeller spinning at close to same speed give a high circular flow , just thinking it may tame the incoming fluid .
guessing about 10 pounds oil and the converter spinning , it's some good fab work and he is having fun it seems with tuning the process.

I like this because I get to learn something new from watching.

Honestly that's what you don't want. You basically want the oil to fall into the spinning canister. The centrifugal effect is how it filters. Think of it from a cross section view. Here's a pic that might help a little. Basically the oil drips into the spinning canister. The cross section of the canister is simply a ] shape. The rotation causes the oil to fill the cross section. The inertia carried by the heavier particles help settle them towards the outskirts of the canister. Allowing the lighter oil to move inwards. Eventually after so much oil is there, it starts flowing up towards the top where it finally is flung out. (Cleaned) If you were to introduce turbines like the converter originally had, you'd loose this ability for the fluid to freely "float" to the top.

If the OP doesn't mind, I'll try to post up some pics of the setup I tinkered with. (I still have it actually) Same principal, much smaller scale.

 

[glassk]

Honestly that's what you don't want. You basically want the oil to fall into the spinning canister. The centrifugal effect is how it filters. Think of it from a cross section view. Here's a pic that might help a little. Basically the oil drips into the spinning canister. The cross section of the canister is simply a ] shape. The rotation causes the oil to fill the cross section. The inertia carried by the heavier particles help settle them towards the outskirts of the canister. Allowing the lighter oil to move inwards. Eventually after so much oil is there, it starts flowing up towards the top where it finally is flung out. (Cleaned) If you were to introduce turbines like the converter originally had, you'd loose this ability for the fluid to freely "float" to the top.

If the OP doesn't mind, I'll try to post up some pics of the setup I tinkered with. (I still have it actually) Same principal, much smaller scale.

Redirecting the flow is not going to help in this application, . would like to see ideas on your inside to trap dirt and the exiting of cleaned oil, maybe in another thread If you don't mind some questions

 

[dikwks]

"If the OP doesn't mind, I'll try to post up some pics of the setup I tinkered with. (I still have it actually) Same principal, much smaller scale."

Absolutely I would NOT mind at all. The more info the better. My major goal here is to make cleaner fuel for my heater. Cleaner reduces wear and allows me to operate it more.

With regard to stuff on the inside to trap dirt etc. , the C-shaped section is the only thing required and the reason for using the TC. It already has a radially symmetrical shape that is already balanced (mostly), and will spin true.

 

[olbucksnort]

I found this company that used a balance induction cone to direct the even flow of incoming WMO. Oh, Great job on your fuge you've inspired us.

 

[glassk]

I found this company that used a balance induction cone to direct the even flow of incoming WMO. Oh, Great job on your fuge you've inspired us.

Thanks, after looking around some I see more of what is being done and marketed, that was my line of thinking that fluid should come in closer to center .

 

[o1951]

I found this company that used a balance induction cone to direct the even flow of incoming WMO. Oh, Great job on your fuge you've inspired us.

I do not see how the sludge comes out. Manually disassembling and cleaning after each batch is not the answer.

 

[tobyS]

Re balancing, I have made several static balancers that work very well. I make sawdust blower impellers with very heavy plate (40 hp or so) and they run very smoothly. I could get a picture or two of a balancer if anyone has an interest (but don't have one here to take a pic of).

They are made from two 5/16" plates turned to about 4" dia with a ball bearing center. Their OD intersects so as to cradle the shaft and they are mounted on a piece of angle to clamp down (there are 2) to a bench with the impeller hanging over.

Heavy spots go to the bottom and the very low friction makes it work to within a few grams. I weld at the opposite side back of the blade, until it does not rotate on it's own. I do the same when putting hard surface on them.

My guess would be, there is a heavy spot that is causing a vibration that a simple static balance would allow you to fix and run at higher rpm.

 

[dikwks]

Thanks for the responses. I LIKE it! Simplecentrifuge was one of the ones I looked at and had determined that would be the one if I decided to buy. I saw the cone shaped insert and also saw the $100 price tag. The way I got the fluid into the bottom of the bowl is with that section of copper tubing you can see in the first set of pictures. The bowl is about half-full so you can't see the end of it. I like that static balancer idea, it may be the way to go if the add weight and test idea doesn't work out. And as an added bonus it would be a good reason to get the lathe. I'm not sure how well adding wheel weights to the inner rim of the bowl will work out as it puts the weights so near to center. In the meantime, I've got a jar of the product setting aside. No evidence of anything settling yet.

Re cleaning: Even the bought centrifuge looks like it would need to be cleaned out manually. I would be more inclined to clean it that way as it lets me look at what is coming out of the oil. Alternatively, I could wash it out with gas in place and take out a screw to drain it into a different container, but that would allow contaminates into the catch basin. Something I'm not too hip with.

 

[tobyS]

We used a De-Laval centrifuge in making bacon bits, real ones, to separate the rendered fat from the solids in a through feed process. I'll look into how they worked (but I don't work there any longer). How would bacon fat run? LOL

 

[o1951]

I like that static balancer idea, it may be the way to go if the add weight and test idea doesn't work out. And as an added bonus it would be a good reason to get the lathe.
.

Well, here is the old man's (me) way of balancing stuff without no fancy equipment. Works good for me for over 50 years.
I spin her up, and lightly touch a crayon or pencil to the outside. Stop it, and look. Where I see the mark is the heavy spot. I either grind a bit off, or put some on other side, and repeat.


Just be careful not to get more than the pencil (your parts) near anything spinning.

 

[tobyS]

o1951, that's how one would make it run true, but it would not balance it. A combination of both methods would not be real difficult. What I described for static balance is certainly no fancy machine, it's simple and can be extremely accurate for equalizing weight. If one only does your method, there could be a large imbalance and it could shake itself apart at high rpm. And centrifuges operate best at very high rpm.

 

[glassk]

Component Specifications
•Standard 3450 RPM Motor:
•1/3 Horse Power
•250 Watts
•60 Hz
•120/230 Volts
•1200 G's
•Custom 6000 RPM Motor:
•1/3 Horse Power
•250 Watts
•50-60 Hz
•120/230 Volts
•Up To 3600 G's

•Centrifuge Bowl:
•22 oz Capacity
•8" Diameter x 2" Deep
•5.5" Opening at top
•3 Drain holes in bottom
•Can accept a booster cone
•Quick-release torque nut



•AC Drive:
•Input: 120/230 Volt Single Phase
•Output: 230 Volts @ 3 Phase
•Fully Programmable
•Not Compatible With GFI Outlets
•Centrifuge Housing:
•13" Diameter x 18.5" Tall
•(2) 3/4" threaded ports on bottom
•(3) 3/4" threaded ports on top
•Snap-lock lid with lid seal
•Throughput: Up To 25 GPH
•Centrifuge Heater Element:
•60° F to 250° F Adjustable Thermostat
•7.25" Long x 3/8" Diameter Thermowell
•(2) 304 Stainless Steel Heating Elements
•7.5" Immersion Length x 0.315" (9 mm) Diameter
•1,000 Watts (39.2 w/in2 Watt Density)
•120 or 240 Volts
•1.25" NPT Thread


some specs I found

and the link for rotor runout

http://www.simplecentrifuge.com/gallery-series-68.html

 

[dikwks]

The idea that one can mark an imbalance with a marker/crayon does have some history. I worked at a shop that had about 5 different methods of balancing tires. One of them used a light to find the imbalance. A switch was mounted such that it was actuated by the lower control arm as the arm moved down following the imbalance. If there were any deflection of the rotor as it spun, the mark could be used to find the imbalance. That kind of thing probably wouldn't work so good here as there is not enough deflection in the system.

3600 g would surely get the job done, but that set-up as described would run better than $1600. I am way too cheap for that. I'll have to settle for the 2000g this one is producing.

I'm likin' that four disc set-up. Would be a good reason to buy the lathe.

 

[glassk]

Centramatic wheel balancers install behind the wheel. Once installed, centrifugal force causes the shot and oil filled aluminum balancing ring to automatically balance the tire and wheel as you drive down the road.

[h=2]Wheel Balancer Function[/h]The free moving hardened lead alloy shot is bathed in a propietary synthetic oil. This two part combination moves automatically to the side opposite of the heavy points in the wheel assembly. The automatic function of these balancers can even compensate for mud, snow, and ice build up with the ability to correct front wheels that are up to 16 ounces out of balance and rear wheels up to 26 ounces.

I was thinking of a tube around the bowl filled with shot, of course if you had a lathe, you could build it in the coupling and then true the end and have stand by bowls for the kids to clean up

 

[tobyS]

glassk, the link has one of the balancers like I make and use in the second photo. I calculated the rim speed of your examples and the 3450 rpm with 8" is 120.4 ft/second and the second, 6000 rpm is 209 ft/second. That will give something to compare to.

dikwks, how many rpm is your shaft, taking into account the motor and pulley ratio and what diameter is it? It appears to be about 14" in your photo. How did you determine your going at 2000G? That seems very high if you haven't balanced it.

As far as marking it with a marker and taking materials away thinking you are balancing it based on eliminating run-out, well, it better be low rpm.

 

[glassk]

When I saw it I knew that was it, the rim is 1.5 inches smaller than the main body of 8 inches, so is it using the diference to push the oil out, because it's a function of RPM and Diameter with the average thickness of fluid in there to. for a low mod rotor a CO2 tank or Fire Extinguisher , rounded to the valve could make the lip end and turn a 5 or 6 inch long plug diameter of tank to slide in and couple to that. then length could determine volume and

 

[o1951]

o1951, that's how one would make it run true, but it would not balance it. A combination of both methods would not be real difficult. What I described for static balance is certainly no fancy machine, it's simple and can be extremely accurate for equalizing weight. If one only does your method, there could be a large imbalance and it could shake itself apart at high rpm. And centrifuges operate best at very high rpm.

Not really - that does not make it run true - what it does is give it a dynamic balance. I realized this loooong ago, watching an on the car wheel balancer, which does the same thing - spins wheel assembly up, heavy spot exerts the most force, causing the most displacement. Operator puts weights on opposite side, spins it up again to check.

The tire/wheel assembly still does not run true, but it is balanced. When they wanted to make it run true, they would shave the tread.

 

[glassk]

bottom is a stand alone temperature controller, will operate a heater to set point and takes several kinds of thermocouplers

top right is your AC drive , and the top left is a 4 to 20 milliamp loop controller, I just have too much stuff

 

[tobyS]

I guess we can disagree o1951.

In the case of the wheel balancers, they have sensors that determine the amount of out of balance condition and record it in relation to the position on the balancer shaft. I think they are rather sensitive load cells. That is entirely different from marking the spot that sticks out the most and reducing it back by grinding. Your method is not dynamic balance, it is marking run out and reducing it. If reducing run out gives a better balanced condition, like "shaving the tread", then luck is on your side. Even tire and wheel combinations that have been made run true by shaving must be balanced.

I'm still hoping dikwks will give numbers for his rpm and diameter to compare.

 

[tobyS]

Glassk...one can never have "too much stuff" if your an inventive type.