Engine Dynos

[drivetrainUW-Platt]

We have the Land and Sea snowmobile water brake dyno for the snowmobile team and I would like to use it for our Formula engine.
The biggest challance is how to configure the drive system since it is made to slide on the output shaft of the snowmobile engine. I will have to get a stub shaft with a keyway similar to the one on the snowmobile but that is easy.
The dyno is not designed to be run at too low of an rpm so I can have any gear reductions in the drive system. This said, a chain drive off the transmision would mean I would need the same gear as on the tranny and the chain would be worked too hard turning those tight corners and would die quickly.
I had thought about a direct drive with a rubber joint like this:
http://cgi.ebay.com/ebaymotors/BMW-E30-318-E36-E28-528-...237275QQcmdZViewItem (http://cgi.ebay.com/ebaymotors/BMW-E30-318-E36-E28-528-Driveshaft-Coupling-Flex-Disc_W0QQitemZ110081237275QQcmdZViewItem)

Also a belt drive like off a car would work. Either way I need to drive it off the transmission output shaft somehow.

If you guys have pictures of yours I would love ot see them. I want to make this all fit on a stand that can be rolled around and fit thru a standard 3' door.
I also need to figure out a cooling system for the engine. I think the dyno runs an open loop just using a garden hose, so maybe I can use the same water supply for the engine????

[VFR750R]

As far as coupling the dyno, I HIGHLY recommend a direct drive. Chains do NOT last, even nice 530 o-ring chains. Plus as the chain wears it effects power levels a measurable amount.

A simple driveshaft with universals works really well, we use it on our cup dynos. With universals, remember to keep a small offset in the two shafts to keep the universals turning. The shafts should be parallel, just offset enough to put 1-2deg of angle in your driveshaft.

I've heard good things about the rubber flex coupler you linked to, but have never used them. Even BMW or some other OEM uses them. In that situation your shaft should be perfectly aligned if possible.

On cooling. I suggest a cooling tower. Basically a tall tube (usually aluminum) where hot water goes in the top and cooled water to the engine comes out the bottom. It's basically a mixer of engine water and cold city water to get the temp water you want going back in the engine.


Place two tubes in the tank. One near but not at the top is your overflow. Peferably a bypass should be put on this to keep some pressure on the tank (5-10psi) A good source is a knuckle valve, which can be bought at a plumbers store.

The second tube roughly 1/3 down the tank is your cold water in. This should have a mechanical thermostat valve attached. The termocouple for it should be place at the tower outlet to the engine.

This tank can be made very fancy where the cold water tube comes in from the bottom and is perforated to help distribute the cold water, but overall it's a pretty simple cheap way to control water temperature.

Forseeing some issues you may have in temp control. There are electronic valves you can buy, which you would also need a PID controller sold by watlow or omega. This is expensive but effective. Another helpful cheap trick is a water pressure regulator to change water pressure going into your tower. This can help the flow of your thermostat valve be more linear and controlled.

Don't forget about your oil system as well. A lot of sustained dyno time will heat your oil quickly, and changes in oil temperature have an effect on power. An idea I had very similar to the cooling tower (which is a proven device) is to basically put an oil cooler in the cooling tower or in its own cooling tower so water instead of air is going over the fins. I'd be willing to bet it's VERY effective in temp control.

[Superfast Matt McCoy]

We use a chain. We have the dyno connected to a shaft with a sproket welded on. You can probably get about 50 dyno runs on a good expensive sealed chain, and if you run it right, you can get two usable lengths out of one chain. If you use a radiator, make sure it's big. We had one out of my old Dodge Dart that didn't really work. Now we have a big aluminum one out of a Dodge diesel or something. You can do it fairly cheaply and make it simple if you want, but I spent much more time last year fixing the dyno than I actually did doing dyno runs.

So, you should probably do an output shaft, cooling tower, and all that other stuff VFR750R suggested.

http://sae.ou.edu/gallery/images/photos/SRT-2006/manufacturing/rb2.jpg

[Alexandre D.]

We use a direct drive with universal joint. It has worked flawlessly for many years. I wouldn't recommend a chain due to the effect on power. I don't know if there's issues with reliability. Of course chain drive may be the best results/time solution.

For engine cooling we use a huge radiator coming from a pickup-van. They are dirt cheap when you buy them used. Something like 3'X2.5'X2". I would recommend PID cooling like VRF750. An on/off system will have a very long time response. The overshoot may affect the engine.

The dyno runs on city water. It's not the most nature friendly solution but it's the easiest http://fsae.com/groupee_common/emoticons/icon_wink.gif

For oiling, we use an F4i with the stock oil cooler and it has never been an issue.
I'll try to make some picture today.

[Buckingham]

GOOD industrial belt drives are a lot less messy than a chain, last longer, add damping, and tend to do a lot less damage when they explode.

[rjwoods77]

About direct drive. We have a 1"OD keyed shaft and I made a nice adapter along the same lines as the one that land/sea sells for .75 shaft for kart engines. The dyno just mount solid on the end of the shaft and when we had it running a 7000rpm(our redline) the housing wasnt even shaking. I would assume that the best way to do things is mount it right to the crank like we do. I dont see why you couldnt do the same off the splined output of the tranny or if it is possible remove a sidecover and mount it directly to the crank snout if ther eis one so you dont have to bother with transmission induced effects. Our old cbr had an adapter that was from a sprint car shop that made one for when you use it front engine longitudal. The key is to keep the dyno as close to the engine as possible when you do this. Ours is about 9 inches from the crankshaft bearing and we have had zero problems with it.

Just curious about water supply. We just hook a garden hose to one end and dump into storm drain on the other end. I havent looked into wether you are supposed to have more pressure(head) or flow into the unit because it isnt my department. Are we doing a stupid engineering student kind of thing or is this acceptable?

[Dallas Blake]

We're more fortunate that most schools in that our school has a superflow SF901 and an SF902. Our Team has taken over the SF902. We ended up making a drive shaft with CV's. What VFR750R said about the CV's is bang on, you want a very small angle in your CV's

For cooling we use a cooling tower and an electric water pump. Our water for the cooling system comes from the city

For oil cooling we run a dry sump so when we are running under a load we have a 24" diameter fan pointed at the sump to cool the oil. So far we havent had any problems

I wouldn't recommend running your dyno off of the crank. Most dyno's dont have the capability of running upwards of 12000RPM. I know the maximum on ours is 10000 RPM sustained and 12000RPM for no more than 30 seconds. Maybe im wrong about the land and sea dyno but i know superflow doesnt like to see more than 10000RPM

I have a couple pictures of ours on my Engine Page main page.

http://engsoc.queensu.ca/formulacar/enginepage.htm

[rjwoods77]

http://www.land-and-sea.com/snowmobile-dyno/snowmobile-dyno-specifications.htm

They say it is capable of higher rpm than fsae will see. I think crank mount is the way to go after seeing other setups in person. Less crap and more robust.

[VFR750R]

On crank or trans shaft mounting, you'll have to look at your engine and how easy it will be. I've seen a slightly shady but effective setup with a sprocket welded to a yoke for the right spline but much safer would be to get a yoke cut for your transmission spline.

Rob is right that it's the more ideal senario. By eliminating the clutch and transmission, you've eliminated variables and extra heating and breakdown of your oil.

Beyond what you guys should worry about for FSAE, but should keep in mind is the crank on 4 cylinders is designed for the power to come off between cylinders 3 and 4. When Honda first did this back in the 70's, it helped torsional vibrations of the crank and they could design the crank lighter after that...

I'd take power off the trans shaft.

[drivetrainUW-Platt]

Rob, lets see some pictures of your crank mount, that sounds kind of harry!

VFR,
Did you guys run your fsae or another radiator with your cooling tower or was it just that?

Do you have any pictures of your universal joint setup?

Thanks guys, getting a lot of good suggestions, keep them coming please!

[Benjlv]

our team using a radiator submerged in water for cooling along with the stock thermostat. For the most part the system holds temperature fairly well to the point that we are happy with in. For oil cooling we are about to do the same with a oil cooler and thermostat setup.

that said i have a question for teams using the larger superflow brakes. We have a SF-90x and have all sorts of problems holding loads. a lot of this problem is our lack of a proper controller, but i am curious what kind of performance other teams get with setups that have proper control systems.

[VFR750R]

Well when I was on the team we used a radiator out of a V6 lumina with a fan, but I don't recommend it. It was not a steady water temp.

We used cooling towers at work, and they are the easiest to setup unless you want the system to be pressurized (above 10psi or so), because the city water has to fight your system pressure and it's hard to allow only enough water to come out to hold your pressure.

The universal setup is just like a pto for a tractor or a driveshaft. The trick way to do it, if your output shaft has a grove for a retaining plate is to have a pin push into that groove just like a tractor pto.
Another way is to have a fixed length shaft, and mount the engine so that it can't come off, but this will require you to simultaneously hook up your driveshaft as you mount your engine.
A driveshaft in a car is easy enough, it slips on the trans output and then straps hold the universal in the diff yoke.

The last way is to have a telescoping driveshaft that you can fix in place once you get the length right. This may be the fastest, but probably the car way is the best.

[rjwoods77]

Mike,

http://fsae.com/eve/forums/a/tpc/f/8356059423/m/87510195831

Here you go. I dont have any actual pics of the setup right now but I will post them when I do. That dimenision is 3.29 inches from the mount face of the engine to where the clutch begins. I didnt have a broach handy that day so we cut a slot and welded a key in there. A little filing inside to compensate for warpage and thats it. It works great. Everything is held concentric and there is absolutely no backlash. Like I said before land-sea offers something just like this but with a broached keyway for .75 shafty cart engines. So for something like a cbr I would to internal splines to match the output of the trans. My crank is threaded for a 3/8-24 bolt which clamps the whle deal together. Do the same thing for the cbr since it has a bolted end. You can hug the unit up pretty close. Something I would like to add though. There was one team that converted their cbr to use a cvt by doing something with a side cover and a shaft extension. Be a cool thing to make up and use instead of going through the tranny and clutch which I am sure shows up on the dyno. Much less wears a clutch pretty good.

[BeaverGuy]

Benji,

Oregon State uses a SF-901. My first year there, we had many problems with properly controlling the engine. However, most of those were issues with the people that were running the Dyno. They didn't really know what they were doing. The next year I read through the manual so I could understand understand all the controlls, there are atleast 3 I can remember that influence the ability to hold a load position. After that the only control issues we had were with major torque spikes and dips due to intake/exhaust resonance. I think more of this could have been dealt with by adjusting the pressure boost valve but we didn't really have time to deal with it as it was really only two spots in the RPM range and one of which was below operating range.

[drivetrainUW-Platt]

Did you get those dimensions for the taper from Land and Sea? It looks like there is a taper in the dyno with threads halfway through?
So with your setup you pretty much just have one shaft straight to the dyno. Do you wish you had some sort of cv/u joints or rubber doughnut couplers for misallignments and dampening?

[rjwoods77]

I called Land/Sea and asked what the dimensions on their taper were. The threads at the end of the taper are for the removal bolt. Adapter slides on the crank. Dyno slides on the taper. A bolt goes through the whole thing and clamps the assembly down to the crank snout. When you want to remove, you place the removal bolt into the threads of the dyno and the end of the bolt contacts the front face of the adapter. Basically the same shit they make

http://www.land-and-sea.com/dyno-accessories/dynomite_adapters.htm

No I dont wish I had a cv because I think it is stupid when you can drive directly off the crank. There are no misalignment or vibration problems. If you are so set on using a cv just do so.

[rjwoods77]

Mike,

Attach this to the front sprocket drive hub.


http://www.speedpartz.com/dwarf_driveshaftadapter.htm

[John Stimpson]

I drilled an 8 hole bolt pattern into a countershaft sprocket, and then machined a "stub shaft" that on one end, featured the required snowmobile clutch taper. On the other end was a flange that had the matching 8 hole bolt pattern on the modified countershaft sprocket. The stub shaft was sized long enough so there could be a pillow block misalignment bearing supporting the moment on the stub shaft.

NOTE: To ensure the countershaft sprocket and stubshaft were concentric, I chucked the stubshaft in the lathe, and then used the live center to hold the countershaft sprocket up against it, and then bolted them together. Of course, CHECK YOUR RUNOUT!

Lastly, you might realize doing this disallows you from being able to install the bolt that holds the countershaft sprocket onto the countershaft... I was wary of this, and checked it very frequently, with zero signs of fretting, peening or wear.

[Boston]

I understand the argument that chains must be replaced every so often on a chain driven dyno, but I don't understand why the power loss through the chain is being considered a negative. When you use the data gathered from the dyno (for vehicle simulation, thrust charts, ect) you would want to account for the power loss through the chain.

That said, we use a direct drive on our dyno. Works pretty well.

[Superfast Matt McCoy]

Originally posted by Boston:
I understand the argument that chains must be replaced every so often on a chain driven dyno, but I don't understand why the power loss through the chain is being considered a negative. When you use the data gathered from the dyno (for vehicle simulation, thrust charts, ect) you would want to account for the power loss through the chain.

That said, we use a direct drive on our dyno. Works pretty well.

The efficiency of the chain is going to change between a 14-tooth/50-tooth sprocket pair on the car and the 14-tooth/14-tooth sprocket pair we have on the dyno (necessary for the dyno to spin fast enough.) If you did use the correct size, you would still be ignoring the CVs and the hopefully negligible bearings in the drivetrain. If you want accurate wheel thrust you really need a chassis dyno, or some wheel speed sensors and a parking lot. Our engine dyno is used for comparative testing of intakes and such, so precision is more important than accuracy and the chain becomes a nuisance, and sometimes, a projectile. It does its job well enough for me to not want to change it, but if I was staring from scratch it might be different.

[VFR750R]

Originally posted by Superfast Matt McCoy:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Boston:
I understand the argument that chains must be replaced every so often on a chain driven dyno, but I don't understand why the power loss through the chain is being considered a negative. When you use the data gathered from the dyno (for vehicle simulation, thrust charts, ect) you would want to account for the power loss through the chain.

That said, we use a direct drive on our dyno. Works pretty well.

The efficiency of the chain is going to change between a 14-tooth/50-tooth sprocket pair on the car and the 14-tooth/14-tooth sprocket pair we have on the dyno (necessary for the dyno to spin fast enough.) If you did use the correct size, you would still be ignoring the CVs and the hopefully negligible bearings in the drivetrain. If you want accurate wheel thrust you really need a chassis dyno, or some wheel speed sensors and a parking lot. Our engine dyno is used for comparative testing of intakes and such, so precision is more important than accuracy and the chain becomes a nuisance, and sometimes, a projectile. It does its job well enough for me to not want to change it, but if I was staring from scratch it might be different. </div></BLOCKQUOTE>

Exactly, knowing how much your chain takes in power is a good thing to know, but adding that variance when testing actual engine parameters is a recipe for trouble.

The same could be said of chassis dynos...I hate em. They add 20+ variables to a system with tons of variables already. The arguement that they simulate real world performance is a fallicy. Water brakes and eddy current dynos are great for that without the compromises made running an entire chassis.

[drivetrainUW-Platt]

If we can find one of those transmisison output adaptors that were posted to fit our engine, I may go that way. I havent seen the case pulled off our engine yet to look how I would mount to the crank.
Still wondering If I want to go driect drive or belt. Driect is sounding simpler, I would just have to find some sort of a bushing/coupler to fit the tranny adaptor to the shaft for the water brake.

[rjwoods77]

I posted a trans adapter a couple posts ago. Use that nad make the taper cone piece to bolt to but make sure you do some sort of hub-centric system to maintain concentricity.

[drivetrainUW-Platt]

Originally posted by Rob Woods:
I posted a trans adapter a couple posts ago. Use that nad make the taper cone piece to bolt to but make sure you do some sort of hub-centric system to maintain concentricity.

exactly...I havn't heard back from that tranny spline adaptor company yet, guess I will have to pay for it, darn!

Anyone else want to contribute to the cooling tower vs bigass radiator cooling system for a dyno.

[drivetrainUW-Platt]

Anyone have a good idea how I can speed up the input to the dyno? The waterbrake isnt designed to operate at low rpms and after going through the transmission might not be useful at all under 5K rpm.

[VFR750R]

I hope someone will reply that has actually run direct. We ran 16 on the engine/19 on dyno shaft tooth sprockets with the engine in 6th gear with no issues. I think we tried in 5th and could just overload the dyno at high boost...but it's been a long time. Either way, it's worse then a direct drive and we did high boost loads as low as 4500.

[Poe]

I've seen a slightly shady but effective setup with a sprocket welded to a yoke for the right spline but much safer would be to get a yoke cut for your transmission spline.

Shady and super sketchy without a doubt. Normally aspirated, we used to run in 2nd gear, but with the turbo, we've done the dyno tuning in 3rd. 5000 rpm is about our lower limit too. The brake we use is quite over-sized for a 600cc motorcycle engine.

[Bradford]

We tried 1:1 out of the trans, no good for us,couldnt hold the engine at all, full load trying to hold the engine back and the engine would pull through the absorber. We put a smaller restrictor in the outlet to get the pressure up and it still didnt help. Ours needed the step up drive.

It depends on what kind of absorber you have, the bottom exit or the side exit. The side exit has more holding power than the bottom exit. We have the bottom exit one.

I was told a hugh rad would be needed by lots of people, but that wasnt the problem, problem was air flow through it.
Used an 85 camaro 2 core rad, nothing special, had it lying around, and an electric fan off of a ford minivan, used only one side of the dual fan set up, could use both fans for more cooling

WOT -- CBR600f41 -- 2:1 step up belt drive from trans output to land and sea dyno, step test 5000 --> 9000 engine rpm 500 rpm increments 10 second holding per rpm, ==> 2 degree water temp change at completion.

I used my heat transfer equations to half-a** the amount of area for radiator and air flow needed then used the biggest things i could find over these values. Worked good for me

Belt drive is a 8 rib serpentine pully arrangement off a ford supercharger drive connected to a harmonic balancer off a ford truck. approx 2:1 step up ratio, working against the trans output 1.822:1 reduction.

custom sprocket to 1" keyed shaft adapter at the trans output -- intermediate shaft riding on two pillow bearings -- large drive pully.

ford serpentine tensioner used to keep belt and stretch in check.

8 rib supercharger pully, slug welded into center then bored out to 1" keyed 1/4x1/4 shaft to fit dyno input.

and a big cage around the whole thing so no one dies if the belt lets go

guys at land and sea said if you can use belt, easier on internals of absorber.

Thats what we did

Bradford

[VFR750R]

Supercharger belt is probably not a bad idea if a direct drive won't work for you. Definitely better then a chain. I think what is being explained is a direct drive to an intermediate shaft, which is a good idea and allows you to greatly enlarge the drive and driven pulleys to get the load down on the belt.

That also brings up the water inlet. To get the maximum holding torque from the brake, you'll need a good source of inlet water. Most teams, I'm assuming, use city water, but a tank with a float valve that a seperate pump sources from can greatly increase the flow and pressure into your brake if your city supply is lower then average pressure or there is a lot of upstream flow restrictions. This is important to keep brake temperatures in check as well. I'm sure land and sea has a maximum temperature the brake is designed to run (our Froude brakes at work are 160F)

[drivetrainUW-Platt]

Our dyno is a side exit port. The snowmobile guys just run city water, but they call them selves "slednecks" (rednecks that ride sleds) so I dont know if I can go by what they have done. I still havent decided on belt of drect off the tranny. Direct is more compact and less components, but alignment might be an issue. With a jackshaft and double belt there is room for speed changes but then you need a lot more parts and room.
Any more pictures of peoples setups?

[drivetrainUW-Platt]

For the cooling tower, its a constant loss system that uses city water right? Some of the team members are concerned that city water running through the engine at operating temperature will leave behind mineral deposits and restrict the cooling passages and cause problems with the water pump......

[VFR750R]

The cooling tower is a constant loss or also called 'open system'.

This can be rememdied by making a copper coil inside the tank so you have a closed system. Then it's basically a homemade heat exchanger. I've never done that so I can't vouch that it will cool enough, but with enough 'rounds' of copper, I can't imagine why you couldn't get it to work with some basic heat transfer math and a couple stabs at it. I've also considered a air to water intercooler boxed off to create a water to water intercooler. Again a little math and you can pick an appropriate size to do the job without making it impossible to control.

For sure the city water will lead to some buildup and corrosion. We have aluminum tanks and they last about 5 years using them 45 hours a week, before getting leaks at pipe thread connections and cracks on internal tubes. I'm guessing no engine you have will see that much dyno service before it's blownup or wore out.

[BeaverGuy]

The dyno I used at Oregon State was a SF-901 with the cooling tower that operates as described with city water. I also opened many of the engines. The only one that I found that had any considerable build up was an engine that had been sitting in the car unused for about 4 years. All of the other engines had minimal corrosion and deposits. I would see no problem with running off of a city water supply unless the water has an unusually high concentration of minerals.

[drivetrainUW-Platt]

That is my thinking as well. Someone suggested the idea of having a tub of water with the radiator submersed in it and constant loss supply into the tub....since water is better at transfering heat then air it might work out.

[VFR750R]

Beware that response time on the rad/tub design will be poor unless executed very well.

I've also thought about taking a radiator and welding and boxing it in with baffles such that it makes it a 4, 6, or even 8 pass on the city water side. That way it will take a lot less water and the temp stability will be better.

[drivetrainUW-Platt]

Could you wip me up a drawing of a cooling tower, just a cutaway in something like paint or similar. I was thinkin that the float shutoff valve from a toilet could be used to control the water lever in the cylinder....

[BeaverGuy]

You might try the Superflow website. They have some of the manuals on there for download and in the 901 manual the have a cutaway of the tower.

Barring that as far as I can remember the SuperFlow cooling tower is controlled by a thermostat. There is a manual valve on the tower that you open to allow water in until it reaches a certain level. Then you turn off the valve. The thermostat allows city water in and hot water out to keep the water in the tower at a fixed temperature. Then there is a manual valve to allow you to drain the tower. A float valve would work for initial fill but then you would need a thermostat to regulate water exchange.