Allwheel Drive Omni GLH SRT4

[black86glhs]

I would seriously doubt that the viscous clutch is slipping when you turn it. They used viscous clutches in the older Bravada transfer cases to allow for AWD and being able to turn. Only downside was if the unit failed it would allow the thing to roll even in park unless you had the parking brake on.

[phantomrt]

^^^ Sounds like something General Motors would engineer. My Jeep Cherokee has a transfer case that allows turning in 4wd. It allows 1 wheel drive while in 4wd, but I'd take that over a vehicle that doesn't have a park gear that'll reliably hold the thing in place.

I just got in from tearing into both my 1992 PTU and 2005 Pacifica rear differential so I can triple check everything. As most know, the PTU's ratio is 1:3.42. The pinion (the small gear) has 12 teeth and the ring gear has 41. Divide 41 by 12 and it comes to 3.416666666666666666666.

The Pacifica rear end has 11 teeth on the pinion and 38 on the ring gear. Divide 38 by 11 and you get 3.45454545.

3.4545 - 3.4166 = .0379. Those are very close ratios. I could sit here and do more math myself, but the service manuals say that the speed difference on opposite ends of the viscous coupler equates to 5-32 RPM, undoubtedly determined by misc factors such as tire sizes and vehicle speed. I'm guessing 5 RPM is at a crawl, and 32 RPM difference would be 80 mph or so.

It looks like the gears in my PTU have not held up well over the last 20 years. I'm pretty sure this thing is going to make some noise. So, I basically paid $150 for a case with the speed sensor hole in it. Boo. I think replacement gears are available from Zumbrota bearing and gear. I may look into that. This would make sense seeing that other people are noticing that the PTU is the first thing to be missing from early AWD minivans when they hit the yard. Its not like these gears are even under any kind of significant load..

[moparman76_69]

So the Pacifica's doesn't have bolt on axles? I've never seen one in person... Seems odd that they would do it that way, with slip-in axles...

This unit uses a torque arm in place of the tube used on the earlier vans (like what I have).

Mike

I posted a picture in post 316. Maybe it didn't show up so here it is again.



---------- Post added at 09:41 AM ---------- Previous post was at 09:37 AM ----------

I just got in from tearing into both my 1992 PTU and 2005 Pacifica rear differential so I can triple check everything. As most know, the PTU's ratio is 1:3.42. The pinion (the small gear) has 12 teeth and the ring gear has 41. Divide 41 by 12 and it comes to 3.416666666666666666666.

The Pacifica rear end has 11 teeth on the pinion and 38 on the ring gear. Divide 38 by 11 and you get 3.45454545.

3.4545 - 3.4166 = .0379. Those are very close ratios. I could sit here and do more math myself, but the service manuals say that the speed difference on opposite ends of the viscous coupler equates to 5-32 RPM, undoubtedly determined by misc factors such as tire sizes and vehicle speed. I'm guessing 5 RPM is at a crawl, and 32 RPM difference would be 80 mph or so.

It looks like the gears in my PTU have not held up well over the last 20 years. I'm pretty sure this thing is going to make some noise. So, I basically paid $150 for a case with the speed sensor hole in it. Boo. I think replacement gears are available from Zumbrota bearing and gear. I may look into that. This would make sense seeing that other people are noticing that the PTU is the first thing to be missing from early AWD minivans when they hit the yard. Its not like these gears are even under any kind of significant load..

Lucky you found the early PTU. I have a 93 and a 01, guess I need to tear them both apart and find out which one is in better shape. I was going to tear it apart anyway and see if the speed sensor gear was still on the shaft. Got any pics of the guts?

[phantomrt]

I did not take any pictures because there wasn't much of anything interesting to take pictures of. Eliminating the teeth for the speedometer gear would yield significantly lower production costs, so I am willing to bet that they are gone. The gears should interchange, however. The ring gear unbolts and the pinion is the typical thing you would find in any solid rear axle setup. The PTU is a very simple mechanism, but it is appearing as though the gears were not of great quality.

[zin]

I posted a picture in post 316. Maybe it didn't show up so here it is again.

It showed up, but if you compare it to those that I posted, they look pretty much the same (from what you can see). The picture you posted has plugs where the axles would be and it appears to be a factory replacement, and often those don't come with everything, so i just expected that you were to re-use the axles, since they wouldn't necessarily be bad.

Mike

[Reaper1]

Probably a dumb question, but is there any way short of custom gears to have the same ratios front to back to eliminate the perpetual slip in the viscous coupler?

[RoadWarrior222]

Fiddle yer tire sizes.

[phantomrt]

I am still a bit confused by how the Rear Drive Line Module (RDLM) works, and I'm having trouble explaining it.

Even if you matched front and rear gear ratios exactly, there is still going to be some "slip" involved. Preferably, that slip stays at some device made to handle it rather than at the tires. It would be similar to any old school style 4wd truck where it is forbidden to drive in 4WD on dry, grippy road surfaces. It causes significant amounts of extended driveline binding, which will affect durability in time. This is going to happen no matter how close the tire sizes are. They are never the exact same.

The RDLM's viscous coupler theoretically doesn't even slip going straight down the road. This Bi-directional overrunning clutch (BOC) keeps that from happening. When the clutch is in overrunning mode, zero torque is transferred through it and the VCU just free-spins with no slip, similar to as if you just removed the propeller shaft from the vehicle and drove it like that. I believe the BOC is in overrunning mode any time that the BOC's input and output (to the VCU) are different, whether the vehicle is going forward or backwards. Once the input speed matches the output speed due to front wheel slippage, the BOC locks up and the torque is transferred to the VCU which now has to handle the torque it is given. So, given this, the RDLM's input always goes slower than the output. So when under braking the front wheels typically lock up long before the rears and this effect will never allow the RDLM's input to match the output for that split second, so you won't get any drivetrain binding as a result. The anti-lock braking programming supposedly keeps this fact in mind.

In theory, if you fiddled with tire sizes to get the input to the RDLM to match the output of it, it is going to cause the VCU to start working unnecessarily and wear it out. Whether or not it'll be enough to cause a problem any time soon is up for the R&D team. Like I mentioned before, there will ALWAYS be a need for slip involved. By design, there *has* to be a slight difference in PTU and RDLM ratios to maintain VCU longevity.

Helpful hint is that the pre-2001 rear ends have an overrunning clutch that only works one way. Although a bit simpler in operation, it requires more gizmos and physical space to make it work in reverse.

[zin]

Probably a dumb question, but is there any way short of custom gears to have the same ratios front to back to eliminate the perpetual slip in the viscous coupler?

Without a center differential, you need that slip/viscous coupler in order to prevent binding.

The rear end doesn't track exactly with the front, which will make the rear wheels attempt to "drive" the front wheels, unless something allows the front and back to move at slightly different speeds. The difference in the ratios allows this.

If they were the same, without a viscous coupler, it would be much like trying to turn a car with a spool or an open diff that's had it's spider gears welded together (poor mans spool).

In short, it's probably not something you'd want to do... But a possible way to go would be to use a Magnetorheological clutch, something you could turn on and off as needed, but otherwise would allow free-wheeling. I imagine if you replaced the silicone oil in the VC and managed a way to wire it...

Or just leave it the way the factory did it, and see if it is really a problem before potentially wasting time and money...

Mike

[RoadWarrior222]

There might be a slight mismatch due to wheelbase difference, so you might want to re-optimise it.

[moparman76_69]

I am still a bit confused by how the Rear Drive Line Module (RDLM) works, and I'm having trouble explaining it.

Even if you matched front and rear gear ratios exactly, there is still going to be some "slip" involved. Preferably, that slip stays at some device made to handle it rather than at the tires. It would be similar to any old school style 4wd truck where it is forbidden to drive in 4WD on dry, grippy road surfaces. It causes significant amounts of extended driveline binding, which will affect durability in time. This is going to happen no matter how close the tire sizes are. They are never the exact same.

The RDLM's viscous coupler theoretically doesn't even slip going straight down the road. This Bi-directional overrunning clutch (BOC) keeps that from happening. When the clutch is in overrunning mode, zero torque is transferred through it and the VCU just free-spins with no slip, similar to as if you just removed the propeller shaft from the vehicle and drove it like that. I believe the BOC is in overrunning mode any time that the BOC's input and output (to the VCU) are different, whether the vehicle is going forward or backwards. Once the input speed matches the output speed due to front wheel slippage, the BOC locks up and the torque is transferred to the VCU which now has to handle the torque it is given. So, given this, the RDLM's input always goes slower than the output. So when under braking the front wheels typically lock up long before the rears and this effect will never allow the RDLM's input to match the output for that split second, so you won't get any drivetrain binding as a result. The anti-lock braking programming supposedly keeps this fact in mind.

In theory, if you fiddled with tire sizes to get the input to the RDLM to match the output of it, it is going to cause the VCU to start working unnecessarily and wear it out. Whether or not it'll be enough to cause a problem any time soon is up for the R&D team. Like I mentioned before, there will ALWAYS be a need for slip involved. By design, there *has* to be a slight difference in PTU and RDLM ratios to maintain VCU longevity.

Helpful hint is that the pre-2001 rear ends have an overrunning clutch that only works one way. Although a bit simpler in operation, it requires more gizmos and physical space to make it work in reverse.

Sounds like to me if this is the case, then the only point the rear kicks is when the front tires are spinning (no traction), and for all intents and purposes the AWD doesn't really do much.

[phantomrt]

The AWD does absolutely nothing until it is needed, which is what you want in any situation other than in a trail riding rig. It remains front wheel drive until that is no longer sufficient to propel the vehicle. The gear ratio difference is 3.45-3.42 = .03:1. The input to the RDLM has to rotate .03 revolutions more than normal before the rear wheels start helping out. I am having trouble sleeping right now because I am sick, but if I figure it out correctly, the transmission's ring gear has to spin .009 revolutions (3.15 degrees) before the rear end kicks in. That is definitely not much. If the front end starts a 1-wheel wonder, that one tire will only have to spin 6.3 degrees before the rear end contributes to the cause. With ratios this close, I would think that a simple U-turn is darn near enough to get the front and rear axle speeds the same. Someone please double check my math on that. I don't even think that 6.3 degrees of tire spin is enough to induce a tire chirp.

I watched about half of an episode of Top Gear where they had some kind of Alpine country minivan challenge. I believe they used an AWD Chrysler Town & Country, an AWD Chevy Astro, and the last was an RWD Ford Aerostar, which I am guessing had a limited slip Ford 8.8 rear end, but don't quote me on that. They threw on oversized off road tires along with some other typical Top Gear modifications and went on with their usual business that they do with their cars. As compared to the Aerostar's rear wheel drive, the other minivans totally cleaned up on the trail. The front wheel spin of the Chrysler minivan wasn't even noticeable. It there were situations where it appeared as though it wouldn't even spin a tire and climb over stuff that the Aerostar could not. It was literally pushing the Aerostar up the alpine trail. To say that the AWD doesn't do much would only be accurate to say on the streets of Phoenix.

If all you did was drive the streets of southern Arizona in a 150 horsepower minivan, the AWD would be quite pointless to own.

I'll have to verify with the 3000GT technical information manual if/when I get time, but I believe that AWD system has identical front / rear ratios and they are linked together with a viscous coupler. Granted, these cars are driven hard, but a failure of their VCU is not uncommon, and seeing how fast-acting viscous couplers are, I imagine that it is partly due to the fact that the VCU is constantly slipping all the time. Per their popular transmission vendors, the VCU's are shot on nearly all transmission cores. In theory, just a lot of highway driving is going to wear it out in time. The Chrysler RDLM's design will prevent that from happening... in theory. But then, one of their transmission vendors told me that with all 4 wheels off the groun, running the transmission without CV axles in the front WILL ruin the transmission.... a statement which is not true. I'll question their credibility.

[RoadWarrior222]

I watched about half of an episode of Top Gear where they had some kind of Alpine country minivan challenge. I believe they used an AWD Chrysler Town & Country, an AWD Chevy Astro, and the last was an RWD Ford Aerostar, which I am guessing had a limited slip Ford 8.8 rear end, but don't quote me on that. They threw on oversized off road tires along with some other typical Top Gear modifications and went on with their usual business that they do with their cars. As compared to the Aerostar's rear wheel drive, the other minivans totally cleaned up on the trail. The front wheel spin of the Chrysler minivan wasn't even noticeable. It there were situations where it appeared as though it wouldn't even spin a tire and climb over stuff that the Aerostar could not. It was literally pushing the Aerostar up the alpine trail.

I'll have to look out for that episode..

Before "Daves Farm" was deleted off youtube, he had a couple of vids up where he was basically mudding an AWD caravan, did very well, pulled a ford explorer backwards on dirt also, in a tug'o'war.

[Reaper1]

I saw that TGAmerica episode. It proved to me that the Chrysler AWD system is VERY good.

I honestly thought that the VC is what took the rpm slip difference. I forgot about the BOC. DOH! That does pose another question however. If I'm thinking of the over-running clutch correctly, then it works exactly like the sprag in an anutomatic transmission. Anyone who as built a drag race tranny that uses a sprag for 1st gear (like the A413 does), that putting too much torque through it will cause failure and they can wear out. The solution some have used is to engage first the same way it is in "low" where tow hydraulic elements are applied to hold it in 1st instead of 1 and the sprag. This takes the strain off of the sprag (at the expense of having to figure a way to time the release and apply of the hydraulic elements correctly so as not to induce an internal bind and slow the shift).

I wonder if I'm thinking about it correctly, and if this is something that is even worth worrying about? I can honestly say I don't think I've ever heard of that individual part failing in the AWD system...usually it's the VC.

[phantomrt]

I'm not too knowledgeable with overrunning clutches. I honestly only tinkered with just one overrunning clutch in my life, and it was in an old Ford C4 transmission. With how it was set up, I would tend to believe that it isn't going to enjoy tons of torque applied to it, but I've done no research on it to see what they can actually do. With the addition of the viscous coupler in series with the overrunning clutch, it "cushions" the torque load and since the coupler actually slips, it doesn't handle ALL of the available torque either. Pending durability testing, I really like the concept and design of it. It was well thought-up. AWD Pacificas are starting to get old, and are not SUPER common, but the rear diffs are selling cheap and there's plenty out there. That is usually a sign of a part that doesn't go bad often. The next step from here would be to see what the front-drive based Jeeps use in their rear ends. I imagine that they could have some kind of limited slip rear end, and seeing that Chrysler's AWD design really hasn't changed much from 1991-2006, iI wonder what the chance is of the 2007+ stuff working. It appears as though the 2007+ Journey/Pacifica/Avenger use the same PTU and also rear differentials. In light of this, the front-drive based jeep stuff may be compatible. These is all shot-in-the-
dark speculation, but it may give more options for rear ends. Do your own research on this. I didn't spend much time with it.

In the Top Gear episode was a Caravan now that I think of it more. They cut the roof off in order to supposedly save weight to float it across the lake, and they called it the "SS Caravan".

I would never guess that a Chrysler minivan would win against a Ford Explorer in a tug of war. The Explorer is probably heavier and at least often have limited slip rear diffs. I never liked Ford's "automatic" control trac 4wd. It seemed to hold up fine, but when climbing a steep snowy hill, you could feel and hear a bumping noise under the vehicle when it would engage that gave it a slight unpredictable feel. Stability on a slick snow covered highway was much better with a Jeep Selec-trac with open differentials all around.

[zin]

These are some of the pictures I got from Carlos... Hope it helps folks out!

Mike

PS I have some others of the driveshaft, etc if anyone is looking for something particular, let me know and I'll see what I have.

[phantomrt]

Note that the caravan differential has a sticker that says "3.45." That must be the gear ratio?? The Pacifica unit I have has no such markings. It is still a Getrag unit though.

The more I study the AWD project, the less intimidating it becomes. One of the toughest items may be relocating the darn gas tank, which will probably end up where the spare tire well is. I would like to keep a spare tire--a compact spare will work just fine--so I am open to ideas as to where that can end up. Just throwing it in the trunk was not my favorite idea. When it is all said and done, I want a car that I can comfortably drive across the country. I want the car to appear normal, but I have a whole bunch of tricks up my sleeve which will make someone who knows these cars make a double-take and wonder what the F*#*@ this and that is. Oh the fun that this will induce. It'll undoubtedly become a couple year project that'll be similar to how one builds a custom hot rod.

[rich tideswell]

Subaru style gas tank?

[Reaper1]

I've seen quite a few modern cars these days that have saddle-style fuel tanks. In most cases they have a balance hose between the sides and/or a sump pump on one side to pump fuel to the side where the main pump is. The return (if it has one) always goes to the main pump. I don't know how the sump pumps are controlled, though. I'd imagine they don't let them run dry when that side of the tank goes dry.

[zin]

Indeed the fuel tank is something that HAS to be changed/modded in some way to fit the prop shaft, Carlos said he just bought another Omni tank off eBay and put it behind the axle... I'd prefer to find a saddle style that would fit, or even get a "spare tire" cell from ATL, etc (they make them for the BMWs that road race), kind of small, but for a cruiser it probably wouldn't be too bad, though there may be gas fume issues since it'll be pretty much inside... but I like th3e idea of a safety cell vs stock tank if it's going to be in the "crumple zone", better still to be outside...

And best yet, finding a suitable saddle style that could be made to fit in the stock location and perhaps even add some capacity (or at least maintain it).

Most of the saddle style tanks will have one "live" pump with a fuel pressure driven "jet pump" that essentially pumps fuel from the "dead" side of the tank to the "live" side, keeping the powered pump in fuel and emptying the tank completely.

I imagine one could build two tanks and connect them with a cross-over pipe with the same jet pump/powered pump arrangement... Might be how I go... Well see if/when I get to that part.

Mike