M220 e-Locker Differential Deep Dive Overview - skip if you're TLDR prone...

[Silver-Bolt]

This solves the rumor that it had a welded ring gear.

Sponsored

 

[Snacktime]

Genius! Proprietary Ratios and welding the ring gears seems pretty well aimed at adding a level of expense and difficulty at the average do it yourself mechanic. I wonder was this Ford directed or vendor?

Ram has been doing this since 2014 on 1500 front differentials, this is becoming standard across multiple makes and manufactures.

 

[mpeugeot]

You could probably use a 4.30, but that's barely close enough. I'd be happier is it was a little closer. But even then, I don't see where there is a 4.30:1 available for the M190 either. What are the 4.27 gears you have from? I'd be interested to see what people are finding and what will ultimately work.

I agree, I have a set of Yukon 4.27's, but they are not reverse cut, so I don't think that they will work. I would have to disassemble the M190 to know for sure. Now, as for the welded on gears, I do have access to a large enough lathe, milling, and grinding equipment to actually forcefully separate the gear from the differential - but that's really going about it the hard way. I seriously don't want to have to resort to that. I would much rather just buy a ring and pinion from Ford in the correct ratio; so I have opted to wait a bit.

I don't have to install it right now, the rear locker and advanced 4x4 have made short work of anything at my level of expertise off-road. Eventually, I will throw the Eaton LSD in there, but that's just because it's a full-time torque biasing LSD and requires no thought or wiring to engage.

 

[Rocketeer Rick]

This solves the rumor that it had a welded ring gear.

No, it doesn't really change anything on that front. We've long said that the open diff has a welded gear, the locker is bolt on. This confirms that the locker is bolt on. That's all...

Now, as for the welded on gears, I do have access to a large enough lathe, milling, and grinding equipment to actually forcefully separate the gear from the differential - but that's really going about it the hard way. I seriously don't want to have to resort to that.

If it were that simple, everyone would do it.
Unfortunately, its not. I'll put aside the aspect of "can you weld it well enough" for now. To be fair, we have built prototypes exactly how you described. But they were by their nature short-lifespan vehicles.The thing to know, though, is that the ring gear weld process is very specific. Both in process and, as a result, design. The interface between the ring gear and the diff housing is entirely different for a welded ring than for a bolted. They aren't simply eliminating the bolt holes and running a bead around the joint. That's not to say you couldn't install a ring gear on a diff and weld around the edges instead of use the bolts, but that is not really what the axle manufacturer does.

The shape of the ring gear is different and the way it butts up to the flange on the diff case is different. The shape of the flange itself is designed specific to promote a good weld joint; this includes the weld contact area, the shape behind the weld contact area, and the venting of the weld area itself. The bottom line is that, even if you machined the 4.27 ring gear off of the OEM diff, the mating features on the TruTrac that you bought are not the same. The ring sits on the diff case differently, the flange is a different size in a different location. They aren't made to interchange, and you can't assume that even though they're both for the same axle model that they will.

I have seen what the Dana M220 weld vs bolted ring version of the diff case look like, and the ring interface is nothing a like. By the same token, Ford uses a version of the Super 8.8 in the Explorer, but has a welded ring instead of the Mustang's bolted. I've designed diffs for both of them that are in production at Ford now. Again, you cannot remotely swap around ring gears on those either, there's too many differences. So I have every reason to believe that the M190 will follow the same pattern.

Now we look into the welding itself - manufacturers that are welding rings to diffs have put a lot of effort into the specific metallurgy needed to produce a good weld. As you can imagine, they use these material specs on the diff cases that get welded. They are not nearly that specific on those that aren't (because: cost), so you might struggle to get a good weld in the first place just due to iron chemistry or pearlite content. Also keep in mind that the gear is a gear steel, probably medium carbon alloy. The casing for the diff is ductile iron. The stress load on the weld joint is a sort of can opener effect fatigue situation, continuously trying to deform as the ring rolls past the pinion mesh. that's a tough set of conditions to work with.

In production, Ford (and I presume Dana as well) laser welds (though I know that GM uses electron beam in some cases) so that they get very predictable control of the size and penetration of the weld bead. They did a lot of development to get to something that worked, especially since the weld is steel to iron. Could you create a weld that would hold up to the fatigue loading long term? Maybe. Could you reweld a gear that's already been welded on, then cut off back onto a new diff case in a way that controls distortion (because you have to maintain ring runout and flatness of that your gear lash is consistent) well enough? I can't answer that, but keep in mind that at this point, the weld area on the ring has already been heat cycled a couple of times so its properties and shape have changed. Odds aren't in your favor even for a skilled welder.

With all that said, I would absolute not tell you not to try it. If you want to experiment, try new things, or even play Dr. Frankenstein with your vehicle, by all means, have it. This is, of course, how we all learn stuff and figure out what is and isn't possible individually and as a community and I would love to see what you learned in the process. My goal here is simply to make you aware of what you're up against in getting it to work.

 

[mpeugeot]

No, it doesn't really change anything on that front. We've long said that the open diff has a welded gear, the locker is bolt on. This confirms that the locker is bolt on. That's all...

If it were that simple, everyone would do it.
Unfortunately, its not. I'll put aside the aspect of "can you weld it well enough" for now. To be fair, we have built prototypes exactly how you described. But they were by their nature short-lifespan vehicles.The thing to know, though, is that the ring gear weld process is very specific. Both in process and, as a result, design. The interface between the ring gear and the diff housing is entirely different for a welded ring than for a bolted. They aren't simply eliminating the bolt holes and running a bead around the joint. That's not to say you couldn't install a ring gear on a diff and weld around the edges instead of use the bolts, but that is not really what the axle manufacturer does.

The shape of the ring gear is different and the way it butts up to the flange on the diff case is different. The shape of the flange itself is designed specific to promote a good weld joint; this includes the weld contact area, the shape behind the weld contact area, and the venting of the weld area itself. The bottom line is that, even if you machined the 4.27 ring gear off of the OEM diff, the mating features on the TruTrac that you bought are not the same. The ring sits on the diff case differently, the flange is a different size in a different location. They aren't made to interchange, and you can't assume that even though they're both for the same axle model that they will.

I have seen what the Dana M220 weld vs bolted ring version of the diff case look like, and the ring interface is nothing a like. By the same token, Ford uses a version of the Super 8.8 in the Explorer, but has a welded ring instead of the Mustang's bolted. I've designed diffs for both of them that are in production at Ford now. Again, you cannot remotely swap around ring gears on those either, there's too many differences. So I have every reason to believe that the M190 will follow the same pattern.

Yes, that is my fear. It's a matter of finding out what is or is not possible. I won't know anything until I can get hold of a spare M190. My idea may be completely impossible to achieve, I accept that this could be a complete fail. However, the Tru-Trac would be a nice addition to a non-locker Bronco that was off-roaded but not used for rock crawling.

Now we look into the welding itself - manufacturers that are welding rings to diffs have put a lot of effort into the specific metallurgy needed to produce a good weld. As you can imagine, they use these material specs on the diff cases that get welded. They are not nearly that specific on those that aren't (because: cost), so you might struggle to get a good weld in the first place just due to iron chemistry or pearlite content. Also keep in mind that the gear is a gear steel, probably medium carbon alloy. The casing for the diff is ductile iron. The stress load on the weld joint is a sort of can opener effect fatigue situation, continuously trying to deform as the ring rolls past the pinion mesh. that's a tough set of conditions to work with.

Yes, I would not want to re-weld an already welded and heat treated unit, but I am perfectly willing to machine it... whether it be EDM or mechanical machining, because the part is unlikely to warp. There is no reason for me to weld on the cases, and that is also something that I would tend to avoid. Now, I have successfully welded A356 cast Aluminum and 7075-T6 Aluminum in the past, and it's NOT easy (both "unweldable"). Cast Iron, is equally tricky if you don't know what you are doing (and I don't even bother with cast iron, because it's that tricky, I have friends who specialize in that).

In production, Ford (and I presume Dana as well) laser welds (though I know that GM uses electron beam in some cases) so that they get very predictable control of the size and penetration of the weld bead. They did a lot of development to get to something that worked, especially since the weld is steel to iron. Could you create a weld that would hold up to the fatigue loading long term? Maybe. Could you reweld a gear that's already been welded on, then cut off back onto a new diff case in a way that controls distortion (because you have to maintain ring runout and flatness of that your gear lash is consistent) well enough? I can't answer that, but keep in mind that at this point, the weld area on the ring has already been heat cycled a couple of times so its properties and shape have changed. Odds aren't in your favor even for a skilled welder.

Like I said above, my goal would be to machine it and mechanically fasten it to the appropriate differential. I see the steel as the material that I want to retain and the iron the sacrificial material that I don't want. I need the ring gear, which as you note, probably doesn't share the interface with the Eaton differential. I can machine things flat with ease, what is much more difficult is to prevent warpage when welding ANY metal. Obviously, laser welding, friction welding, and even electron welding can all be potential techniques to accomplish that, BUT they all have risk as you correctly note and all require extremely talented people who know much more about welding than I do.

With all that said, I would absolute not tell you not to try it. If you want to experiment, try new things, or even play Dr. Frankenstein with your vehicle, by all means, have it. This is, of course, how we all learn stuff and figure out what is and isn't possible individually and as a community and I would love to see what you learned in the process. My goal here is simply to make you aware of what you're up against in getting it to work.

I get what you are saying, and yes, this isn't something trivial. It's not bolt up. If I were to weld things in a differential, I would be relying on some friends with decades of experience welding (and blowing auto/aircraft parts up). My goal is to find a way to accomplish this without welding. At the end of the day, an M210 may be the right answer, along with swapping the M220, with the appropriate gears. That said, I still want to have a Torsen style differential in the front and a locker in the rear.

 

[BDrez]

Has anyone heard of having to reprogram or flash the ECU after swapping to a different gear ratio? In other words, if I go from the OE 3.73 to the OE 4.70, is there anything that needs to be programmed?

 

[Rocketeer Rick]

The short answer is yes, you would probably want to do this. Way back when my dad bought a new 1995 F-150, he had the truck's program updated when he changed tire size (same effect as a gear change). At the time, it was to recalibrate the speedo.

That aspect still applies today, but it also gets a little more complicated. I've been told that the PCM is looking at wheel speeds versus engine / transmission speeds. On automatics, they do this in part to verify the transmission is in the gear that the PCM thinks it is. If the gearing changes, it will get pretty confused. If it faults, it might not be willing to drive if it thinks there is a problem with the trans.

That part might not apply to manual transmissions (not sure), but the other factor is the ABS system. The computer is looking for wheel speeds to be within a given range for the road speed. I don't know if this directly applies here, but on my Mustang, if the wheel speed (as a result of regearing) isn't near what it thinks it should be (from the speed sensor on the transmission), the ABS will get confused and fault.

The answer to all these issues is to reflash the computer with updated gear ratio and tire size info. I'd imagine that this can be done with ForScan, as with previous Ford models.

 

[Rocketeer Rick]

On a related note, I just got my hands on an M190 axle assembly from a Ranger. So I'll be digging into that in the next few weeks.

 

[Vikingnene]

Gentlemen,
New to the forum and was hoping I could find some answers. I blazed a trail not knowing what I would come up against, but here is where I am at and what I know so far: I have a 2021 Bronco 2-Door Base manual trans. I won't go into all the mods, but where I have run into issues with the electronics is with tire size and with axle ratio. I've got a set of 37" tires on there and they were causing my tranction control system to freak out, especially on tight turns. Keep in mind this is not because they are 37" tires, as many of you have been running them without any perceivable issues, but that is because most people upgrading to 37" tires are coming from 35" tires and the change is not enough for you to notice the traction control system being a little more sensative. When you go from 30" tires to the 37" tires, however, you notice it. A lot! Any tight turn, even out of a parking spot, and it starts grabbing brake. Hook it up to FDRS and record what is going on and you find it does not like the calculations between the inside and outside tires during a turn, as the circumference of the tires come in to play in the non-linear equation. This is fairly easily resolved with ForScan as long as that is the only mod you have done. But, in the case of my little 2-Door, I swapped out the axles with Ford Performance axles front and rear. The rear axle is a simple one (M220 to M220) but the front one gave me a few challenges. Going M190 to M210 is something that can be done, but you have to read the fine print now available (not available when I ordered my front diff last June) that states you have to order the front M210 diff with the 4.46 gears if you are converting from an M190 diff. It turns out that the 4.46 version uses a different ring carrier that works with the passenger side axle shaft from the M190 diff., but the 4.70 version is offset and does not allow the axle shaft to seat all the way in to the center carrier without bottoming out on the pins. Time to machine the factory M190 axle shaft, of course! After that was done and all the parts fit together properly, it was time to re-gear the front and rear axles. As many of you know, there are no gear sets available for these, so when I found a combination of gears/bearing/seals/crush sleeves that worked back in February, I was ecstatic! Install the axles with the new 5.38 gears and I am almost 100% back to my original gear ratio with the 30" tires and the 4.46 gears. Enter the electronic nightmare!!! I have been battling this for quite some time now and this is where I have gotten to: The BCM (Body Control Module) will reprogram (using ForScan) to the 5.38 gears. From there, the AWD module and the PCM both freak out, so you have to program them as well. Easier said than done... The AWD module is fairly simple, as you can tell it to relearn it's parameters from the BCM and all should be good. The PCM, however, is not happy. I theory it works the same way as the AWD module for a relearn, but it errors out saying the new axle ratio is outside parameters allowed. And without the PCM onboard with the other modules and them being on the same network, the AWD receives contradicting signals from the PCM compared to the BCM and throws a code and the PCM throws a code for having a different value than the BCM. For the icing on the cake, it allows you to drive normal for a short period of time and then puts you into limp mode (i.e. very limited throttle and no boost, to the tune of flat out on the freeway at 50mph). Makes for a very interesting drive! Now, you can pull over, turn off the vehicle, let it reset and then start in on your journey again only to have it repeat that 3-6 miles later depending on how you drive (I have found it to be RPM dependent). I know it took me a long time to get to this question, but does anyone know what the parameters are for the PCM (Power Control Module) as far as axle ratios are concerned? And if not, is there a work around for this issue? A way to fool the modules? The transmission output shaft sensor is obviously not happy with the calculations when compared to the axle shaft sensors (wheel speed sensors). Any and all information would be appreciated. I'm at the end of my rope and really would prefer not to pull the axles back out only to re-gear them again with the factory 4.70 gears (which I only assume are within specs for the PCM). Someone please have a cool hack for this...

 

[Rocketeer Rick]

Vikingnene, I think your post would get more attention if you started a new thread to ask about this. It seems like someone must have dabbled in this already, but if no one sees your questions because they're buried here, you won't get very far.
Also, please try to break that up into some paragraphs or something. A huge block of text is hard to read and people will skip it instead of trying...

 

[Just_Jim]

Technical writing, no, that's not it specifically. I mean sometimes its part of the job, but I'm a development/application engineer that pretty much specializes in differentials, axles, drive systems. So, I'm more on the design/development side of things. In this case, I happened to get a hold of one of these for benchmarking study; what I posted was adapted from my own report on it. But I'm glad you guys dig it.

I'm a little confused here, so maybe you can help clarify something.

We know from Ohm's law, a circuit will only draw the needed current relative to the load and it's resistance. By saying the ECU sends 5A and then reduces that to 2A after a given interval, does that mean the controller applies another resistor into the solenoid circuit to consume some of the available current?

 

[Rocketeer Rick]

I'm a little confused here, so maybe you can help clarify something.

We know from Ohm's law, a circuit will only draw the needed current relative to the load and it's resistance. By saying the ECU sends 5A and then reduces that to 2A after a given interval, does that mean the controller applies another resistor into the solenoid circuit to consume some of the available current?

I don't know the ins and outs of the controls. I'd have assumed that they're stepping down the voltage so that the available amps are reduced. But again, I'm a gearing guy not a double-E. I'm sure there's a body control module that's running the show, though.

 

[Just_Jim]

I don't know the ins and outs of the controls. I'd have assumed that they're stepping down the voltage so that the available amps are reduced. But again, I'm a gearing guy not a double-E. I'm sure there's a body control module that's running the show, though.

I can see it both ways - drop the voltage via a resistor in-line, or drop the current via a load in parallel. I guess the real question is just how much voltage is needed to keep the spring compressed after the locker is engaged?

 

[Rocketeer Rick]

Well, there's a little more to it than that. There would also be separation forces between the lock plate and the side gear, since the face spline has a slight pressure angle to it. Putting a torque across that will result in a thrust trying to push them apart again. Couple that with the vibration in the driveline just from operating, and you have several factors that are trying to disengage the lock. But your point remains valid.

However, that's why it has the tapered ramp / cam holding mechanism. If its drawn into lock, and one tire has a want to spin (or you even turn slightly) and the tapered cam on the lock plate will engage the tapered holes in the casing. This creates a counter-thrust force that will absolutely overcome the return spring force, lock plate separation, etc. So, in practice, the coil only needs to stay on for the times that the holding cam gets momentarily unloaded again (turning back the other way, etc). Also, typically, a closed magnet has a lot more holding power than one with an open air gap. So in the engaged position, it probably is generating a lot of force without needing as much current.

 

[Just_Jim]

Looks like the common solenoid saver circuits drop voltage, so that's probably the ideal way to go.

Many thanks for your quick and insightful reply. I'm one of the guys doing our own Sasquatch conversions and am at the point of wiring up the lockers.

Sponsored