JMortensen Posted July 10, 2007 Share Posted July 10, 2007 Hey, do you think we could get you to pull apart the Quaife when it shows up? I was able to find numerous threads that say that they're identical to Quaifes internally, but none actually showed the two diffs apart side by side. I was also able to find several places that talked about replacing the crappy springs that they come with. You might not want to rule that out as a fix yet. Unfortunately I've looked and I can't find any pictures like we had on that other thread where they show force vectors and all that for the Quaife or the Torsen that works basically the same way as the Quaife (can't remember if that is the Type I or II). Here's another bit to continue the discussion from last night. If you just wanted to spread the side gears apart to create friction on the case, you wouldn't need all the spiral gears. You could go as cheap and easy as a Phantom Grip setup with two big plates that fit between the side gears with springs in between to force the gears apart. But that's not really what the gear driven units do. Its true that the side gears will get PART of the load, but by using the spiral gears you can really reduce the loading on any one part by spreading it out on many surfaces inside the diff. The side gears force the spiral gears out against the case, and so the friction is created by the tips of the spirals rubbing on the inside of the case as well as the ends of the spirals rubbing on the springs, and the spirals meshing with each other. The drive from the side gears to the spirals, and then their hitting the inside of the case DOES in the end push the side gears as well, but that is not the primary friction producer. In fact in the torsen diff, both side gears actually move in the SAME direction, Either they both go to the right or both to the left. I did find this on zhome.com and it explains the OTHER type of helical LSD with force vectors and all that stuff and has pictures of all the gears and how they move. The big difference between this one and yours is that the spiral gears are held on shafts. This forces them to the end of the shaft and it is then ONLY the ends of the gears and their teeth that produce the friction, where the Quaife and OBX and the other style torsen don't have any shafts for the spirals to ride on, so they actually move side to side (move into the springs) and also move away from the side gears until they come into contact with the case, where the end of the gears (toe) can drive against the inside of the case itself. http://zhome.com/ZCMnL/tech/Torsen/Torsen.htm Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 10, 2007 Author Share Posted July 10, 2007 You don't need any testing to see that they are of very poor quality, next to that if you google search on OBX you see that on other forums they also change out the washers. with success. however good luck with it.Share on what you find and not call bullshit on! the qaife part uses a similar set up untill than ill be happy motoring with belville washers Well, yeah, you do. Steel on steel with no pressurized oil cushion between them makes for a fast wearing bearing surface that will gall quickly and eat itself...period. I don't care what kind of steel it is, if you stack 2 pieces of the same metal next to themselves, then press them together with a lot of force and spin them against each other, they will be destroyed. Especially when there is such a tiny amount of contact area between the components. Flatwashers would fair much better because of the increased contact area, but would still be destroyed over time. Gears don't have this problem because by nature of the shape of the meshing gear teeth, the oil on the surface is more or less pressured locally between the meshing teeth, protecting them. The material might be inferior, but that is not the root of the problem. Putting better washers in will only prolong the agony...something else needs to be redesigned. I've ordered up a pack of bellville washers, shim washers, and torrington bearings from McMaster. I'll attmpt to reassemble this thing properly so that it won't eat itself....ever. I'll post some pics after the parts come in and I figure out the correct stackup heights. Quote Link to comment Share on other sites More sharing options...
JMortensen Posted July 10, 2007 Share Posted July 10, 2007 You don't want to put torrington bearings in there because you'll reduce the friction and the torque bias ratio. Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 11, 2007 Author Share Posted July 11, 2007 The only friction I'll reduce will be under decel, which I don't care about. The washers are not supposed to be a friction surface anyway. The only purpose that they were designed to serve, is to preload the side gears outwards against the case, which they will still be able to do properly. The bearing will only allow the washer stacks to rotate against each other without becoming mangled....at least thats my theory. Quote Link to comment Share on other sites More sharing options...
JMortensen Posted July 11, 2007 Share Posted July 11, 2007 The only friction I'll reduce will be under decel, which I don't care about. The washers are not supposed to be a friction surface anyway. The only purpose that they were designed to serve, is to preload the side gears outwards against the case, which they will still be able to do properly. The bearing will only allow the washer stacks to rotate against each other without becoming mangled....at least thats my theory. Quaifes and OBX's are 1 way diffs and act just like an open diff on decel. The springs have nothing to do with deceleration. I guess my arguments haven't convinced you of that though, and I seem to have run out of internet available proof like the Tru-trac exploded view and diagrams that were on that old thread. Oh well, I tried... Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 11, 2007 Author Share Posted July 11, 2007 Quaifes and OBX's are 1 way diffs and act just like an open diff on decel. The springs have nothing to do with deceleration. I guess my arguments haven't convinced you of that though, and I seem to have run out of internet available proof like the Tru-trac exploded view and diagrams that were on that old thread. Oh well, I tried... I agree with you that they are designed as 1 way diffs, in which case you must agree with me that the washers are ONLY there to preload the side gears outward against the case. The washers are not intended to act thrust surfaces under decel, but they do, offering a bit of LSD function, intended or not. With the torrington bearings, the washers will still provide the same or better preload on the side gears, but will have no frictional qualities. Quote Link to comment Share on other sites More sharing options...
JMortensen Posted July 11, 2007 Share Posted July 11, 2007 No sorry, don't agree with that. I see the springs as performing much the same function as the belleville springs in my Nissan CLSD. Although they have an effect on preload that is not actually their main function. Their main function is to make the diff perform more smoothly. You can preload the non-sprung Nissan diffs just as easily and effectively as the sprung diffs (detailed info here): http://www.gordon-glasgow.org/lsd1.asp Since I used to sell the Tru-trac and get complaints constantly about how it made crunching noises when turning, I can understand why the springs are added, and I think it is specifically to prevent those noises. As I've said before, I had a model Tru-trac sitting on my desk for months and I have put some time into understanding how they work, so I'm pretty confident that I'm right in this case... Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 12, 2007 Author Share Posted July 12, 2007 I recieved my new washers and other parts today, finished disassebling things, cleaning, and began reasembling. I noticed a few things. 1: I was wrong as to why the washers were destroyed 2: I still don't have any idea as to why the washers were destroyed 3: I longer care why or how this diff design works 3: there is a lot of room for error in the assembly of this differential. Here are what the original washers looked like when I removed them. Here is one side of the diff case, with one of the side gears and the corresponding 6 helical gears dropped in. There is a left and right side gear, and six left and right helicals. If they are installed on the wrong sides of the diff, the helicals will push the side gears against each other instead of outward against the case. I'm not sure how they were installed the first time. This is how far I tore the unit apart to reposition the washers the first time around. This is the spacer that goes between the left and right side gears. These are the bellville washer retainers, and another shot with one of them dropped into the spacer Here is the spacer, with 4 washers installed between the two retainers. There is a lip on the OD of the retainers that catch the washers, so that the function of the washers is only to push the retainers apart. I was wrong about the washers being a friction surface....I didn't have the diff in front of me and forgot the exact washer retainer mechanism. It appears now, that the washers, spacer, and retainers are an isolated spring mechanism inside which abolutely no movement can possibly take place. So I started measuring to make sure I wasn't jumping the gun on my assumption. If I build the diff and install only the side gears and spacer (no washers), there is 4.2mm of clearance between the stacked assembly and the cavity in the case. This means that in order to have 0 preload, and 0 clearance, the washers must be stacked in a manner so that the retainers stick out of the spacer by 2.1mm on each side of the spacer, and take up the 4.2mm of clearance. The spacer measures 22.95mm thick 1 retainer measures 10.90mm thick 1 retainer measures 11.10mm thick, but the spacers are not perfect. There is about a 0.1mm thickness variation in each one. Lets call the average combined thickness 21.95mm The lip on the OD of each spacer is 3.0 mm wide, and located exactly in the center. The total available space for the stackup height of the washers and the retainers needs is 27.10mm to achieve zero clearance. Each bellville washer has an unloaded height of 2.31mm, and a material thickness of 1.55mm. To acheive zero preload, we need 5.64 washers stacked up like ()()(). 6 washers then would be the minimum necessary for quiet operation. This will make the space between the lips on the retainers 12.45mm, and consequently the space between the retainers will be 4.45mm. but we're not using 6 washers, we're using 8, and with a material thickness of 1.55mm, all the washers have to be fully compressed and flat in order to bolt the assembly together. (since the lip is 4mm from the edge). Since the material thickness of the washers is 1.55mm, that means there is nothing to keep the 2 center washers () in place and centered....and they can fall out and contact the splines inside the spacer, then get crunched as the gears smash toward the spacer and force the retainers back flush with the surface of the spacer. This may be what killed my last set of washers, it makes sense because only the 2 washers in the center of the stack were completely garfed up.....but I'm still not sure how it could happen. This may be why the washers were originally stacked )((()))(, so that all of the washers in the middle kept themselves aligned and couldn't fall out of place and get smashed. If I stack 6 washers up ((())) the stackup height is 11.06mm. If I add the next 2 )((()))(, I end up with a stack up height of 15.6mm. This puts us well within the goal of the 12.45mm minimum washer stackup height required. If the two washers on the end are flattened (they will be the first 2 to flatten), the total uncompressed stackup height will be 14.16mm. Now, here's the kicker. With all the washers fully compressed and flat, the stack up height is 12.45mm, exactly the same as our available space....so there is no room for the retainers to move inwards, and therefor no way for the washers to get smashed. WTF am I missing here? The only possible scenario that I can think of is the cheap ass bolts that are used to hold the case together are stretching when I put the power to it, causing the case to spread apart a bit, and the side gears to move further away from the spacer, allowing the retainers to move apart and the spring washers to relax and fall out of line. Then when I get out the throttle, the whole mess comes smashing back together and the washers get impacted to death every time I get off throttle. I don't think the carrier bearing preload would allow this to happen though....is it possible that I'm flexing the whole pumpkin? I dunno though, its a long shot. any ideas? I'm going to read through all this jargon and probably edit it a few more times, so if you need clarification on anything....please ask. I'm getting confused myself at this point. Quote Link to comment Share on other sites More sharing options...
Flexicoker Posted July 12, 2007 Share Posted July 12, 2007 I just took a real quaife apart about an hour ago. It had six belleville washers stacked: )()()( I think it needs that arrangement so that the wide part of the washer is seated against the washer retainer. If it were ()()() then the 'point' of the washer would go through the lip on the retainer. I aligned them with my pinky, then put the retainer on top. Having them covered in gear lube will help keep them in place. I also had it in a vice oriented with the stub shafts running up and down. Then when you put the stub shaft in there should be less than 1/4 or so of the side gear showing above the element gears. If theres more than your washer stack fell over and is stuck inbetween the retainers. I beleive that if the diff were manufactured with a spacer instead of the belleville stack, it would still work as intended IF and only if it were manufactured perfectly and there was no slop at all. So the spring stack is taking up the slop in the middle. It seems to me that the belleville stack and retainers, and the spacer are allowed to 'float' side to side. So that while differentiating the whole center spring assembly and the stub shafts move side to side. This is going to cause more friction against the endbells on one side and less on the other, and this is where the torque biasing comes from. I heard that the ONLY way to change the torque bias ratio is by changing the angle of the helix on the gears. Thats one incomplete theory, my other, more believable theory is that those are black magic helical gears. Soaked in a potion of elfs blood and rosemary while someone chants 'beetlejuice, beetlejuice, beet....' Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 12, 2007 Author Share Posted July 12, 2007 I agree with what you are saying, but when the unit is assembled, all of the washers are completely flat, so there are no points.....and there is a 4.45mm gap between the 2 retainers.....the washers are only 1.55mm thick, so that means that 2 washers are not supported at all, and 2 more are only about 1/2 way supported. My diff uses 8 washers, that when the diff is assembled, are compressed flat. I've been doing my reading, and I agree with you that the only way to change the bias ratio is to change the helix angle on the gears...in which case the only purpose served by the washers is to keep everything preloaded so it doesn't clank around. I can make a spacer... 6.25mm long, and use only two bellville washers on each side of it in the retainers. that would completely retain both washers, and each retainer would catch almost 1mm of the spacer, ensuring that everything stays where it was supposed to. The diff still worked fine with the destroyed washers that were in it (no pre-load), so I think it should still work fine if it had half of its intended preload due to only using 4 washers, or better yet, make the spacer 9.35mm long and use 1 washer on each side....then that would even cover some case bolt stretch. Waddya think? Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 12, 2007 Author Share Posted July 12, 2007 Thats one incomplete theory, my other, more believable theory is that those are black magic helical gears. Soaked in a potion of elfs blood and rosemary while someone chants 'beetlejuice, beetlejuice, beet....' I think you may have figured it out. Maybe I'll sacrifice my cat on the workbench next to the diff while doing a chant of some type, and it'll start working. Thats probably what they do in england with the Quaife.... The reason the Chinese one doesn't work is cuz the workers probably ate the cats for lunch before they put my diff together, and the gods of speed can't understand their chants Quote Link to comment Share on other sites More sharing options...
JMortensen Posted July 12, 2007 Share Posted July 12, 2007 It seems to me that the belleville stack and retainers, and the spacer are allowed to 'float' side to side. So that while differentiating the whole center spring assembly and the stub shafts move side to side. This is going to cause more friction against the endbells on one side and less on the other, and this is where the torque biasing comes from. I heard that the ONLY way to change the torque bias ratio is by changing the angle of the helix on the gears. That sounds just like the Torsen info that I linked to. The Torsen info said that the side gears both moved in the same direction. The Ford "Goldtrak" differential is a gear driven unit that you can actually buy different helical gears for and they have differently pitched helixes, some more aggressive than others. I don't think that this is the ONLY way to stiffen up the diff though. John Coffey sent his Quaife out to EMI and they preloaded it differently to stop his from spinning the inside tire in long sweepers. And like I said before, some of the Trachtech Trutracs have a bolt and wedge type thing that you can tighten to adjust the preload. I think what you're basically doing there is taking out the slop in the side gears, which allows less (possibly no) movement before the spiral gears start to hit the case, which is what creates the friction and limits the slip. Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 12, 2007 Author Share Posted July 12, 2007 So, Whats the vote on the washer/spacer situation? Quote Link to comment Share on other sites More sharing options...
Flexicoker Posted July 12, 2007 Share Posted July 12, 2007 I think that your spacer setup could maybe work, with )spacer( If the sidegears don't get closer to each other during operation. I don't think they do... but I'm not sure? One other problem I could see is that less washers are going to increase your spring rate, pretty drastically if you're taking out 2/3 of them. So you'd have to account for that when you figure out the length of your spacer. If you you 6 washers instead of 8, they wouldn't be compressed flat. If you had them stacked )()()( would the center 2 washers have enough gap between their bases to engage the retainers? I think thats whats happening in our quaife. Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 12, 2007 Author Share Posted July 12, 2007 If you you 6 washers instead of 8, they wouldn't be compressed flat. If you had them stacked )()()( would the center 2 washers have enough gap between their bases to engage the retainers? I think thats whats happening in our quaife. Well, if I do the math, the The uncompressed stackup height would we the same, with very little preload, but on Decel, it would still smash all the washers flat and let them fall out...When the retainers are flush with spacer, there is 8.9mm between the lips on the retainers for the washer stack to go. Each washer material thickness is 1.55mm, and 1.55x6 = 9.3mm. Even with 6 washers, the retainers would bottom out on the washer stack before the side gears hit the spacer. Quote Link to comment Share on other sites More sharing options...
Flexicoker Posted July 12, 2007 Share Posted July 12, 2007 I.... don't know. I don't think that the stub shafts move in enough to fully compress the stack. Our shafts have the CV cup right up against the housing, so they can't pull inside the case more that 0.010-.020" or so. In that case you'd be fine with 6 washers as they won't get squashed flat, and they'd still stay in the retainer. I think. Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 12, 2007 Author Share Posted July 12, 2007 I'm still concerned because even uncompressed, the retainers only catch 0.3mm of each washer. When they're compressed, nothing will hold them in place. I'll pull my quaife apart tonight to see how they do it...more pictures to come. EDIT: the quaife (real one) just arrived ::drool:: Seeing these 2 next to each other, now I understand why the Quaife is 4 times the price. The first difference I noticed, is that while the OBX uses 11 8.8 strength M8 bolts to hold the assembly together, the Quaife uses 12 12.9 strength M10 bolts to hold the assembly together. Its also double drilled for the ring gear, there are 20 holes...10 for the M10 bolts and 10 for the M12 bolts. Its a very nice piece. The Quaife also uses 6 washers instead of 8. Pictures to come later. Quote Link to comment Share on other sites More sharing options...
NZeder Posted July 19, 2007 Share Posted July 19, 2007 Question for you re the problems you had and I found out. First the OBX centres are designed for S13/S14's so it is designed for the M12 not M10 as stated above - easy fix. Next issue I have found is the input shafts C-Clips. The earlier long-nose R200's have the C-Clips held in the side gear inside the diff. The OBX centres don't have the clips installed. So this started to get me to think - maybe the S13/S14 are designed like other diffs and to have the C-Clip installed on the shafts. So this lead me to research these cars diffs. Then I found a pic on this site with 300zx input shafts next to S13 input shafts. See attached pics. So can any one confirm if the S13/S14 do indeed have the C-Clips on the input shafts not in the centre. If this C-Clip design has changed as I believe then I will need to modify 300zx input shafts so the C-Clip groove is deeper this way the C-Clip will be retained on the shaft (My diff guy has installed the OBX into the diff so I can't pull out and disassemble to see if the OBX C-Clip groove is a shallow one (it looks like it when looking down the input shaft hole)) So I am thinking that if the C-Clip is designed to be the other way around then if you don't make the mod to the input shafts (to retain the C-Clip) I am guessing that there is a chance the input shafts will float in and out of the centre - maybe enough to slam into the belleville washers = result damage to the washers. That leads to this question - what did the end of your input shafts look like? Were they polished like they were slamming into the washers? I will be taking my input shafts to the diff guy so he can confirm that they will need to be modified. Quote Link to comment Share on other sites More sharing options...
lbhsbZ Posted July 23, 2007 Author Share Posted July 23, 2007 NZeder, the OBX that I had appeared to use the C-clips captured in the side gears....at least thats how I installed it. It wasn't much fun getting the clips out of the old one and into the new one, but I did it, and it worked. The axles still popped out like they were supposed to, so if you are using axles that capture the ring on the axle splines, you should think about how to either modify the diff or the axles, because I don't think you'll ever be able to get the axles out once you install them. The axles themselves weren't hitting the washers, axles were hitting the washer retainers. The problem was that the OBX washers, once compressed, didn't return to their conical shape...so they offered no preload on the assembly. Preload is what holds the 2 center washers in place, especially with 8 washers installed. With 6 washers, they are all held in place by the retainers. Quote Link to comment Share on other sites More sharing options...
NZeder Posted July 23, 2007 Share Posted July 23, 2007 cool thanks for that - so looks like I will have to remove the centre once again, disassemble and install these C-Clips. So what is the general rule then 6 good washers vs the 8 not so good ones or 8 good washers for extra preload? Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.