FE Power Forums
FE Power Forums => FE Technical Forum => Topic started by: jayb on March 13, 2016, 09:16:42 PM
-
I'm in the middle of making a few changes to my race car, and thought that this one might interest a few people. Last year the car was running way to much RPM, about 7700 through the traps during Drag Week, where power peak on the engine is around 6600-6800. I wanted to bring the RPM down some, so the first step was to change out the 4.29 gears for some 4.11s. In addition, the street tires I'm running are only 28" in diameter, and with the radiused wheelwells they look just a little small when they are on the car. So, I'm upgrading to a set of 30" tall tires.
These two changes taken together make it practical for me to drive the car on the street without the Gear Vendors overdrive. So, along with these changes I decided to remove the GVOD to save some weight, plus add an aluminum driveshaft to save a little more. The transmission is an ATI Superglide, but when I bolted the normal tailshaft housing onto the trans in place of the GVOD unit, one problem immediately became apparent: The ATI transmission tailshaft housing has no provision for a speedometer.
With as much street time as this car sees I didn't want to be running around without a speedometer, so I spent a little time investigating my options. I have a complete set of Autometer gauges in the car, including the speedometer, and when I checked Autometer's web site I found that they had an electronic speedometer available in the same style as my other gauges. After downloading the directions for the electronic speedo, I figured out that I could use the same kind of magnetic sensor as a pickup for the speedometer as I'm already using for cam and crank sensors on the engine. So this looked like a pretty good fit.
The byproduct advantage of the whole issue was that I would get a driveshaft speed sensor installed in the car. This could work with the speedometer, and the output could also be datalogged by my EFI system. A driveshaft speed sensor is really a nice sensor to have, because it lets you determine if the car is spinning the tires at the track, and also gives good information to help you dial in your clutch or torque converter. I've been threatening to install one for quite a while now, but there was going to be a lot of fabrication required, so I hadn't gotten it done. This weekend I decided to do it.
My business background is in magnetic sensors, and I know that the gear or "target" that the sensor has to pick up is critically important for good performance. The sensors that I'm using, which are Cherry GS1001 Hall effect sensors, give a suggested gear tooth profile in the data sheet, so I used that as a starting point and designed a six tooth gear that would clamp in place around the driveshaft yoke. I wrote a CNC program to machine the target in two parts, and then after cutting them out of cold rolled steel I drilled and tapped them so they would bolt together around the yoke. Here's a picture of the target:
(http://fepower.net/Photos/Posts/SStarget.jpg)
Next I had to design and build a bracket that would hold the sensor. Attaching it to the chassis didn't make a lot of sense to me, because the transmission mount is rubber, and the air gap between the sensor and the teeth of the target is pretty small, around 0.060". The logical place to mount the sensor was with a bracket attached to the tailshaft housing of the transmission, and fortunately for me this is a machined section on the ATI Superglide. I made up a drawing of what I wanted then machined the first part of the bracket out of aluminum. As often happens with my first time machining programs, a couple of minor machining errors crept in, but the part was still usable; picture below:
(http://fepower.net/Photos/Posts/SSbracket.jpg)
Note the raised portion on one of the bracket halves; I needed to machine another small piece to bolt to that, and hold the sensor. The sensor and the cable I'm using are shown in the photo below:
(http://fepower.net/Photos/Posts/SSsensor.jpg)
I like this sensor because the cable screws onto it; in case of a sensor failure, it can be disconnected from the cable and easily replaced. Also, the head of the cable has two LEDs built into it. One is a green LED that indicates the sensor is powered. The other is a yellow LED that lights up when there is no tooth present in front of the sensor, and goes out when the tooth is present. This makes it really easy to diagnose any sensor problems (or problems with the cable, for that matter).
The body of the sensor is 12mm in diameter (about 0.470"), so I designed a bracket that would bolt onto the first one I made, and had a hole in it for mounting the sensor. Here's a pic:
(http://fepower.net/Photos/Posts/SSbracket2.jpg)
Finally I assembled the whole thing on the car, while adjusting the sensor position for the desired 0.060" airgap. The mounting appears very solid. I have not yet wired it up, but I don't expect any issues with making it work. Here's a picture of the whole thing assembled on the car:
(http://fepower.net/Photos/Posts/SSinstall.jpg)
This whole project took about 14 hours over the weekend, and is probably overkill, but I'm confident that it will survive in the vehicle. I'm anxious to try this setup out in the next month or so, when I start driving the car again. I will report back with any issues I find - Jay
-
Nice work!
-
Nice work Jay
-
I just installed an aftermarket powerglide tailhousing, it had a roller bearing for the yoke and lacked the speedo. I thought it had a boss tho that could be drilled/tapped.
Anyhow, now that you have a speed sensor on the driveshaft, have you considered using it to sense tire slippage and possibly change timing to retain traction?
-
That's exactly what I plan to do at the track, although I think I will be focusing on suspension adjustments to eliminate any tire slippage first, before any timing adjustments. Between this sensor and a Go-Pro looking at the suspension under the car, I think I ought to be able to dial in the suspension pretty quickly...
-
Makes sense.
Tho with a proper setup on the sensor, it could sure make you dangerous and surprising on street tires :P
-
Jay, always the innovator! 8)
Just watch, his next step will be a algorithm linked to the driveshaft sensor and the EFI controller that, by sensing tire slip, instantly kills some timing or like a stutter box, drops some cylinders to kill power!.
-
Good engineering! Overdone is just right for reliability IMHO. I might try that sometime on my dragster using my "CNC" (Chris Noodling the Controls) I would just comment that the Falcon ate the speedo cable around 1976 or so. All those "contest of speed" things on Trinity Blvd I think. 8) I just chunked it and drove with the tach until I lost the motor sometime in 1980. That was daily driving to and from work. Actually, It's a street/strip car now and - it still doesn't have a speedometer.
I've been having some fun with 1/4 aluminum sheet what with motor plates and such. I thought "lets try this" and bought a 1/2" bi-metal blade for my 10" C-man wood bandsaw. I don't cut much wood but boy that saw will carve AL now like a boss.
-
could it double as a rev limiter just in case the tranny or rear breaks? that would be cool so as too not take the engine against a distributor limiter.
-
Between this sensor and a Go-Pro looking at the suspension under the car, I think I ought to be able to dial in the suspension pretty quickly...
You beat me to it. I was going to suggest the use of a GoPro. The Hero 4 allows for some very nice slow motion video, and that can be extremely useful for suspension diagnosis. I'm beginning to become a little bit of a GoPro addict since my Son bought me one to use on my car. And nice work on the speed sensor.
-
could it double as a rev limiter just in case the tranny or rear breaks? that would be cool so as too not take the engine against a distributor limiter.
I don't think that would work since it only senses driveshaft rpm. The motor could be at 10,000 rpm while the driveshaft is only reading 40mph. And it would still have to limit the motor RPMs by way of some sort of ignition cutout.
-
Jay, does the sensor also dictate the number of teeth in addition to the profile of the target gear? If not, how did you arrive at that?
Thanks,
Bruce
-
Between this sensor and a Go-Pro looking at the suspension under the car, I think I ought to be able to dial in the suspension pretty quickly...
You beat me to it. I was going to suggest the use of a GoPro. The Hero 4 allows for some very nice slow motion video, and that can be extremely useful for suspension diagnosis. I'm beginning to become a little bit of a GoPro addict since my Son bought me one to use on my car. And nice work on the speed sensor.
Yes, YouTube has some vids of a drag car's suspension and how it reacts. Pretty cool stuff. I just wish we had same years ago while racing when all we had were video cameras someone held near the starting line to record the car coming off the line. Btw, this was very helpful but not nearly as good as seeing the suspension's movement from under the car.
-
Hemi Joel probably had an old Direct Connection unit lay'n around.Remember those?
-
Jay, does the sensor also dictate the number of teeth in addition to the profile of the target gear? If not, how did you arrive at that?
Thanks,
Bruce
I wanted to put as many teeth on the target as possible, because the sensor can handle frequencies of up to 10 KHz (10,000 cycles per second). At 8000 RPM, the driveshaft is spinning 133 times per second, so 6 teeth is only 800 cycles per second. What limited me to 6 teeth is the requirement to have the target in two halves, so it would bolt together and clamp on the driveshaft yoke. The bolts would get in the way with more than 6 teeth on the target, so I settled on that. That number is actually larger than some I've ready about; for example the driveshaft sensor sold by FAST uses two magnets on the yoke, essentially. It probably works fine, but I figured more resolution was better so I went with six.
-
Jay, always the innovator! 8)
Just watch, his next step will be a algorithm linked to the driveshaft sensor and the EFI controller that, by sensing tire slip, instantly kills some timing or like a stutter box, drops some cylinders to kill power!.
Those are already out there Bob. Some guys are using microprocessors to watch the driveshaft speed sensor for any sudden, non-linear changes to the acceleration, and using that as an electronic trigger to change the tune on the fly during a pass. In fact that is the way an anti-lock brake system works. The microprocessor watches the output of the wheel speed sensors and when there is a sudden deviation from a linear speed or acceleration, it kicks in the ABS brakes to modulate brake line pressure.
-
Yeah, a fella on the FE forums had one some time back in an AC Cobra. Seems that car would be the most appropriate usage for that system, as I can't imagine ever being able to get good traction with some of the monster engines people stuff into those.
-
Those are already out there Bob. Some guys are using microprocessors to watch the driveshaft speed sensor for any sudden, non-linear changes to the acceleration, and using that as an electronic trigger to change the tune on the fly during a pass. In fact that is the way an anti-lock brake system works. The microprocessor watches the output of the wheel speed sensors and when there is a sudden deviation from a linear speed or acceleration, it kicks in the ABS brakes to modulate brake line pressure.
LOL - from a friend that is a senior Ford tech at a Lincoln dealer: Lady brought in a car that would start crapping out somewhere close to 60 MPH. Run fine up to a point, then act like someone threw out the anchor. After several days of testing sensors and head scratching, they figured out that she had a blowout on a recent trip and the replacement tire on the right rear was a bit smaller than the others. Once the wheel speed deviated enough from the others, computer thought the car was going into a skid and applied the brakes.
-
Same with my wifes 500 sl . Over 90 km/h
you could not lock the cruicecontrol and some other
strange things.The shop manager on Mercedes laughed
and said I bet you put on bigger rear tires right?
Yeaaa but..??? Change back to smaller and its like new
:-[ :-[ :-[
-
My cars have working speedometers but I use my gps for a speedometer.
It is mounted up just above the dash where I don't have to look down at the speedometer.
Kind of like having a heads up display.
One of my cars has the tach in the way of the speedometer anyway. ;D
Also they sell gps speedometers that look just like any other gauge.
-
I am once again reminded of my recurring thought that JayB is neater than spats on a duck.
KS
-
i know its different but my son needed a speedo,he mounted a sensor on his brake caliper and a pick-up on his wheel to get his mph
-
Jay, do you think there is enough clearance that the reluctor will not bottom out against the back of the tailhousing? Also, do you think the reluctor will cause any imbalance to the driveshaft?
Overall a very neat and ambitious project. :)
-
I ran through the suspension travel and had plenty of clearance. However, in order to take the driveshaft out, I will have to remove the target; the driveshaft will not slide far enough forward with the target bolted on to allow it to get past the rear end yoke and come out.
I can't imagine there's enough weight there to cause a balance issue, but I guess we'll see...
-
put the bolt heads opposite each other, if it's not too late.
-
I would echo Shady's advice. Hopefully you made the same part in two halves so the bolts are indeed opposite each other.
Driveshafts are very sensitive to imbalance. All you need typically is one or two worm-screw hose clamps to balance a buzzy one. With your gearing that shaft has to be spinning up pretty fast.
Have you also determined the critical speed of that shaft? There should be a max rpm spec for any aftermarket driveshaft and you can run up against that pretty easily in a drag car. Now that the GV unit is gone your shaft is much longer, reducing critical speed.
- Bill
-
Good point on balance Bill.
My uncle Bill had a near new shop van that one day seemingly picked up a bad vibration at 65 or faster. The dealership had looked it over a number of times, wheels, driveshaft, etc. with no luck. Then, after more carping, Dodge sent out a factory rep to look it over. Next day, Bill got the call to come and pick it up. He and the tech hit the highway and all was fine. Back at the dealership, the rep. showed Bill out back in the parts department what it was.
A tiny dent on one side of the very long aluminum driveshaft! Apparently, the local quick oil change shop used a side lift hoist and the lift arms pinched the shaft just a tad. Btw, Bill was amazed such a tiny dent could throw off the balance so much.
-
Balance is a funny thing on high-speed rotary machinery. In grad school I built a laboratory for studying composite helicopter tail rotor shafts. The drive unit was a 1-ton variable-speed motor with a step-up transmission attached. It could zing the output up to 10,000 rpm if we wanted. The whole unit was mounted to a huge cast iron table grouted into the floor
There was a bit of detectable vibration in the base plate when we ran the thing up. The professor rigged up a pair of Hall-effect sensors to look at 12:00 and 3:00 on the output coupling flange and we studied the signal with an oscilloscope. We calculated the correction mass and angular location on the output flange and I machined one up.
My machined part ended up being a 5-gram brass washer! I mounted that little guy to the correct spot on the flange. The 1-ton machine then ran as smoothly as could be at all speeds.
-
You guys are scaring me about the balance. I did not machine the parts so that they had screws coming from the opposite sides, so there is going to be some imbalance there. Saving grace (hopefully) is that it is fairly close to the center of rotation, so hopefully it won't make any noticeable vibration. I guess if worst comes to worst I can have the driveshaft rebalanced with the target bolted on, and then mark it somehow so that it is always installed in the same spot. Or maybe just balance the target...
Critical speed for this driveshaft is over 8500 RPM, so I'm not too worried about that. It is a 4" diameter aluminum driveshaft, and even though its a foot longer than the old one, according to the Mark Williams critical speed tables I won't come close.
-
Jay even though it is close to center I would still reconsider. I would be concerned with premature wear on the output shaft bearing and seal. I don't have driveshaft experience but I do work for a centrifuge company. Forces in a g-field are easy calculation, you may want to run some numbers.
While I don't design the equipment I do balance and service them. As mentioned balancing is a matter of grams only degrees of difference or separation. Also symmetry is always key with design as well as locking fasteners or designing fasteners that are not perpendicular to I guess what you might call the center. Think of how you use the natural forces to keep the fastener in place.
As for locking fasteners we use either loctite, locking patches or locking washers. Typically not split lock but instead nordloc washers.
Keep in mind these centrifuges while balanced are subject to vibrations and the size and speed typically generate g's in the neighborhood of 1400.
-
I dont think you will have any problem
with the balance. As you say close to
center, fairly symetric machined
so I dont think there would be any problem
-
You guys are scaring me about the balance. I did not machine the parts so that they had screws coming from the opposite sides, so there is going to be some imbalance there. Saving grace (hopefully) is that it is fairly close to the center of rotation, so hopefully it won't make any noticeable vibration. I guess if worst comes to worst I can have the driveshaft rebalanced with the target bolted on, and then mark it somehow so that it is always installed in the same spot. Or maybe just balance the target...
Critical speed for this driveshaft is over 8500 RPM, so I'm not too worried about that. It is a 4" diameter aluminum driveshaft, and even though its a foot longer than the old one, according to the Mark Williams critical speed tables I won't come close.
Sorry Jay didn't mean to scare you! Still, you might want to take that shaft into a shop that can check the balance now rather than later. That or make a few progressively quicker runs locally while solely checking for any shake-n-bake!
-
you can always balance the drive shaft the poor mans way, (my way). I put the car on jack stands or a lift, take a hose clamp, install it on the front of the shaft behind the yoke, mark it's location & run the car in high gear. then turn theclamp a little at a time to find the sweet spot. takes a little time but usually works as the screw acts as the weight. nice thing it is not permanent should you change things in the future.
-
With all the pesky little problems discussed that go with forward mounting of the gear or "target" : complex driveshaft removal, limited teeth on the target, tight clearance issues, potential difficulties w/balancing etc I wonder if you considered going the OEM route and mounting the target on or near the rear axle flange as the car mfgs do for their ABS systems?.....
Heck a trip to a pull your part junk yard could potentially yield a decidedly larger target with lots of teeth, matching sensor, mount and even heat shielded wiring etc for all of $10.00.
-
Modern ABS systems use a target on each wheel, not anywhere on the driveline. I guess I could have gone the route of putting it on the rear end yoke, but I've seen them on the driveshaft yokes before so it seemed like the logical place to put it.
I have a balancer for balancing model airplane propellers. I think I might just take the target off and try to balance it independently using that, if it gives me any vibration.
-
It's all good, Jay. You're not going to hurt anything in the short term. If the car starts buzzing down the road you'll know what the problem is!
-
Any way it wont be like when i worked at a tire shop
This guy with a El Camino came in and wanted two
tires mounted on the car. He had retreaded them at
a truck tire retreader. He said he had to be in the
Next town in an hour, So i dont had time to balance
them. I told him to come in next day for balancing
He came in and said there was some shaking in
the rear end. Yea we balance that no problem ::)
I put the wheel in the balancer and hit the go button
And all hell broke loose, It ripped the balancer from
the concrete floor all four 10 mm bolts The balacer
Jumped around and tiped over hitting the El Camino
denting the door while i tried to hit the brake button
:o :o :o The customer running out of the shop
;D ;D ;D ;D ;D Yeeaaa right "SOME" shaking in the
rear end. When i finaly got it balanced it was 1/2 kilo
about one pound out of balance :o and he drowe that
for 60 miles
-
Lars,
Lol, that must've been one big chunk of lead weight! :o
I wonder how you got it to stick to the rim!
Bruce
-
The biggest clamp on wheights half the rim
inside and out ;D
-
Jay, if you used 4 bolts, two on each side, you could position them outside the plane of the teeth, but this would have required more machining from a thicker piece of material. The bolts would not interfere with the teeth that way.
-
A pound of weight: I wonder how you got it to stick to the rim!
SIMPLE
Just go to lowes and buy a bag of concrete and a couple bricks!
-
Or chrome metal tape.Just like on the Direct Connection units.No balancing needed :)
-
have you ever gotten a large chunk of clay stuck on one side of a wheel ? more than a little vib.
-
Jay, if you're concerned about the balance, use aluminum bolts, they hardly weigh anything.
I have my own balancing story. I put a 455 into my 67 Olds 442 that was basically a slightly freshened up junkyard motor out of a car with an automatic. I put a steel manual tans flywheel on it.
The first time I ran the motor while I was actually sitting in the drivers seat, I noticed a distinct vibration, that changed with engine speed. I wanted to make sure that it was not from an engine miss, so I did a power balance test by pulling one spark plug wire at a time and checking for constant RPM drop. That came out fine, so then I Checked the exhaust temp on each header tube with a handheld infrared thermometer. It seemed that all 8 where hitting fine. So I pulled the fan belt to rule out fan, etc. No change noted.
Then I dug out the old flex-plate that had come on this motor, and sure enuff, it had a big ol' counter weight welded to it. I put the flexplate and a damper on the crank out of the old motor ( which was sitting on the work bench) to check the orientation of the flex plate weight in relation to the damper weight, and they were about 15 degrees apart.
THen I looked at the flywheel in the car, and balancing holes where drilled in about the same location as the counterweight on the flexplate. Weird. The flywheel is drilled with a few different clutch mounting patterns, and those holes are threaded all the way thru the flywheel to the front. So I used one of the unused holes to bolt 5 ounces of lead the front of the flywheel, in the location where the counterweight on flexplate was. After checking for clearance, I started the motor and it shook like a belly dancer. So then I took the weight off and moved it 180 degrees and started the car. It was way smoother, but still vibrated a small bit. I experimented with weight untill at 3.25 ounces, the motor was smooth as a babys behind. THen I made up a 3.25 oz. steel weight that I carefully ground to fit the contour of the flywheel to lock it in, and bolted it on with a locktighted grade 8 bolt. I think it is safely there to stay, and it works perfectly. I can't beleive I balanced my own motor without removing it from the car!
-
I don't think there would be a balance issue if mounted on the rear end yoke?
http://www.bing.com/images/search?q=driveshaft+speed+sensors&view=detailv2&&id=493B772912A048D2E32A739F54D54AD999DA8B18&selectedIndex=19&ccid=IT4S9KrW&simid=608040621275614477&thid=OIP.M213e12f4aad61ebe267bb2f5b37d4f1ao0&ajaxhist=0