Home Built Tranny Flusher

[Bill Ellis]

There have been some ideas here for a home built tranny flusher. Here's one that would use two auto oil pumps coupled so one runs as a motor to drive the other acting as a pump. This is possible since gear pumps and motors are actually the same thing. By using the same pump model, the amount of fluid pumped into the trans will be the same as that pumped out so there will be a one-for-one fluid replacement. Using plastic waste baskets as the tanks, standard plumbing parts and two used auto oil pumps, this would cost about $50 - $70 and just might work. Adapter plates will have to be made for each pump outlet, use the pickup tube for the intlet.



So, perhaps some of you guys who really know about this stuff would critique this design and, maybe even build one. If ya think it will work, I will probably build one myself since I like doing that sort of thing.



[Broken External Image]:

 

[Todd Z]

That setup looks like the schematic from an ATS 1000 tranny flush machine.

Looks good,

I just wounder if all you really need is some type of non positive displacement pump on the "back to tranny" line and can simply use a catch bucket on the "from" tranny line.



Or even a gravity fill back to the tranny...

Todd Z

 

[Bill Ellis]

I thought about gravity feed back to the tranny but had some concern that the trans pump might pump out faster then gravity feed would refill. If I was sure that gravity would work then there would be no need for any pumps. Certainly using matching, positive displacement pumps is the sure way to go and used oil pumps ought to be fairly cheap.

 

[Wes Davis]

I have no clue on this stuff but what about using two of those pumps you mount on the end of drill motors. I have used them many times to drain water beds. They are also very inexpensive. There are also several pumps for fish tank and pond applications that could be a possibility also inexpensive.

 

[Bill Ellis]

Hey Me, thanks for the input. The plan, as shown, uses coupled positive displacement pumps; i.e., for one rotation of the pump, a specific amount of fluid is transferred. With such a design, the amount of fluid pumped into the trans equals the amount pumped out one-for-one. You don't need a pump to drain the trans, it's own pump does that. One of the pumps in the drawing acts as a hydraulic motor that runs the "fill" pump". Each volume of fluided going through the "Motor" causes the "Pump" to return the same volume back to the trans. The bypass line allows one to completely fill the return line before and get the air out of the system before the flush cycle is started.

 

[Tim Reynolds]

Just to be a devil's advocate, but what if the the two external pumps operate at a much different flow rate than the internal transmission pump. Would this cause problems? Is it possible for the return pump to operate at a higher rate if there is no restriction and simply drains back into the pan while the other pump might have to fight what I'm assuming will be a slower flow rate of the internal pump?



Does that make sense?



Tim

 

[Bill Ellis]

In the design the flow rate of the "pump" is determined by the flow rate of the trans pump. For each ounce of fluid flowing through the "motor", an ounce of fluid is sent back to the transmission. The "motor" is not working against any pressure since it dumps to the "used fluid tank". The "motor" rpm is simply a function of the trans pump flow rate. For instance, if the "oil pump" transfers 2 cc per revolution and the trans pump outputs 1000 cc/minute then the motor will turn at 500 rpm. The "pump" will turn at 500 rpm and send back 1000cc/min. As Todd Z said, this is the basic design of a commercial tranny flusher. I never saw their schematic but, having some knowledge of hydraulic systems, I thought this would work well.

 

[Wes Davis]

I have to go back to school. :wacko:

 

[John Zuber]

Bill, it just may be easier to "drop the pan", install a drain plug (and clean the filter), refill, drive a short distance and drain/refill one more time. its not a flush, but its better than no maintenance.

 

[Bill Ellis]

JohnZ, I can't do that in my driveway! I'd make a terrible mess. Besides, there is no intellectual challenge in that. I like to design and build things. Boats, telescope, furniture, photo studio equipment and what not. Besides, just found out the pumps I could use are only $13 each new at Advanced Auto; Melling M55 pumps used in medium block Chevy engines. Heck, for about $50 could build the whole thing. For me, that $50 would be worth every minute of doing this project just for the fun of doing it.

 

[Richard L]

The simplest kind would only need a cylinder tank capable of holding about 10 quarts of new transmission fluid with a piston and a few large "O" rings. The piston and inside of the tank should be about 4"-6" in diameter and the pistion should be fitted with grooves to seal the piston against the cylinder side walls.



4" PVC pipe with screw on caps would work just fine (you would need to calculate the length needed to permit a 10 quart capaciity) You could probably get a shop to chuck up a piece of aluminum in a lath and make a piston and cut the grooves for the O-rings so that it fit in the pipe with just enough clearance for the O=rings. The piston could be made in 2 or three pieces just so long as it was tall enough to prevent it from cocking in the cylinder.



As the old fluid is pumped into the top of the tank it pushes the piston down forcing the new fluid out the bottom at the same rate and pressure as the fluid is pumped in.



When the piston hits the bottom, you have changed all the fluid that can be changed (about 98%). Reconnect your cooling lines to the transmission cooler and your transmission is ready to go. Now drain your homemade flusher cylinder containing the old fluid into the appropriate receptical for recycling and your ready for your next transmission flush.



No pumps, no switches, and no valves needed. It runs off the hydrolic pressure from the transmission's internal pump and only has one moving part, the piston. The only maintenance required would be to clean the cylinder after/before each use and the periodic replacement of the O-rings.



...Rich

 

[Bill Ellis]

I thought about that too. 12 qts in 6 in ID pipe would require 2 ft of pipe. Need caps, etc. The problem is getting a proper piston. I don't know how much one would cost and don't know who I would ask to make one. I suspect it would be more than $26; the cost of two M55 oil pumps. Another problem is being able to see when the cycle is complete. Plastic pipe from the hardware store is opaque. Clear acrylic pipe is not readily available and is expensive. It would be interesting to price both alternatives because the pipe method is certainly less complicated.



By the way, the schematic above does not have any switches and runs off the hydraulic pressure of the trans internal pump.

 

[Richard L]

Bill,

A simple rod connected to the top of the piston and extending through top cap of the cylinder could be marked in increments to indicate when the piston has moved down 10 quarts, or 12 quarts, etc.



While I suggested having the piston made at a machine shop out of aluminum, it could easily be madeat home out of a few layers of 1/4" Lexan plastic glued together and turned on an inexpensive wood lathe or even a drillpress. You could easily make two 1/2" thick disc that are about 6" in diameter out of a 12"X12" piece of 1/4" Lexan plastic for less than $10. I can get a pack of 4 O-Rings for about $8, and one or two small O-rings to seal the shaft of depth guage.



You did not show any electrical wiring in your diagram, but since you used electric pumps, I assumed you would be including switches in the wiring, or at lease, so alligator clips.

Also, if your system runs off the internal transmission pump, why do you need electric pumps? I can only assume to avoid gravity feed back to the transmission, and pumping the old fluid to the discard container.



...Rich

 

[Dreman]

The pumps aren't electric. He is using standard Chevy oil pumps that are usually driven by a Chevy engine. He's using one pump in reverse. The pressure from the transmission line, powered by the internal pump in the transmission, makes the first pump turn backwards, making it into a motor driving the second pump. The second pump sucks fluid up from a bucket full of fluid and pumps it back into the transmission. Since he is using two identical positive displacement pumps, so for every quart out of the transmission there will be another quart pumped into the transmission.



Just a thought. I don't know how the Chevy oil pumps are driven, but I do know a Ford oil pump uses a 6-sided drive shaft just like an allen wrench. It would be easy to connect two Ford pumps together by just sawing off the curve of an allen wrench, or just using a Ford pump drive shaft.

 

[Bill Ellis]

Went and bought one of the M55 oil pumps and am now figuring out how to plumb the thing. Fittings and epoxy ought to work. The M55 pump has a round shaft with a slotted end. I plan to couple the shafts with a piece of hose which will also act as a u-joint so that minor missalignment won't be a problem. The hardest part of this is getting the tranny connection fittings. Inverted flare fittings are not commonly sold in most auto and hardware stores. I might just use 6" pipe for the containers, capped on one end with a fitting.

 

[Richard L]

Chevy oil pumps?? Why? That really sounds jury rigged and is a serious point of failure. With two pumps it increases the complexity and the chances of failure.



The reason I don't think the oil pumps will work is because the pressure coming out of the transmission is not that great and I doubt that it would be sufficient to power an engine oil pump, plus you would need to prime both pumps. When rebuilding an engine you need to fill the pumps with vaseline so the pump will even begin to pick up the oil in the pan.



Bill, I'm sorry but I think your idea is too complicated and the engine oil pump operating from the fluid pressure is just not going to work. I wish you luck and applaud your ingenuity, but I see some serious flaws with your idea. I think electric pumps would be better if you insist on using pumps.



The problem is that only part of the pressurized fluid is diverted to the cooler. The fluid pressure diverted to the cooler is not as great as fluid that goes through the rest of the transmission which can reach 65-85 PSI. I doubt that the fluid in the cooling lines exceeds 10 PSI and it might be significantly lower?



When I had my transmission flushed, the flush machine had no pumps moving the fluid. The tech had to run the engine at about 2000 RPM with the transmission in neutral to get enough pressure to move the fluid, and even then it took about 10-12 minutes to transfer the fluid. That's only about one quart per minute.



That flush machine is where I got my idea from, except the fluid tank had a clear glass/plastic window so you could see the old fluid filling the cylinder above the piston and the fresh fluid getting pushed out the bottom. The hoses were also clear so you could easily see the fluid color going in and out of the cooling lines. When the 10 quarts where pushed through, the color of the fluid in both hoses were the same clear bright red color.



The beauty of using a cylinder and pistion is that 5 PSI going in equals 5 PSI going out and only involves one moving part. That keeps everything simple and safe.



...Rich

 

[Bill Ellis]

Thanks for the vote of confidence. It's very possible that what I am doing may not work; but, tinkering around is a hobby of mine. The oil pumps have very little friction so it will not take much flow to turn the pump. The pumps will be primed by the tranny line and gravity so that should not be a problem. As far as complexity goes, it's far less complex than installing amps and head units. In fact, if you can measure, drill holes and use epoxy then it's an easy task. It's true that the flush machines are simple; but they also cost over $1000 for all that simplicity. If my machine don't work then so be it. I'll find something else to tinker with.

 

[Richard L]

Bill,

I am a tinkerer too, so I understand your motivation and your need to "Try it". By all means, Do it and see if it works. If it does, you will have greater confidence and satisfaction that you had an idea and made it work. If it doesn't work, you will still probably learn something and that ain't so bad either.



I am just as guilty. Often times my first ideas are too complicated or over-engineered. My first LeverLift worked very well but had a bit of over-engineering because I thought that was the only design that would work. Later I discovered that simplicity led to reliability, easier manufacturing, lower cost, and better profit margins even though the selling price was reduced.



If kyou get it to work, you will always have the option of making it simpler and better, or simply moving on to your next challenge.



...Rich

 

[Bill Ellis]

Rich, you've discovered the most difficult part of the designer's job......simplicity while meeting all the design specifications. We are plagued with over-complex designs in the aircraft industry. For example; where I work, redesigning major structures of an Airforce transport plane have reduced costs over $50 million per airplane over the past four years. Combining several parts into one. Eliminating excess fasteners. Specifying lower cost, yet, just as good materials. It's easy to make things complicated and I'm doing that on my project; it's already over budget. To some degree I'm constrained to use whatever I can find at the local store. To have something custom built would cost even more. No doubt as I go I'll see how I could have done it for less cost.

 

[Tom Schindler]

Why not let gravity do the work?



Place the tank on top of a 10' step ladder and let gravity put the oil in. When the system pumps it out, let it flow into a bucket.



5 Gallon buckets would work. Since there are 4 quarts of liquid in a gallon, a 5 gallon bucket will hold 20 quarts.





Tom