Map63Vette
Well-Known Member
Very nice, that looks plenty robust. What material thickness did you use?
Finally get to pretend I am one of the cool kids - Holley kit swap
- Thread starter DionR
- Start date
-
DionR
Well-Known Member
Thanks!
The upper crossmember vertical parts are 0.187", the bottom of the upper crossmember is 0.119" and the top is 0.059". The lower is all 0.187 except the mount itself which is 0.25".
The top of the upper has to be hand formed so I didn't want anything too thick. Just wanted to duplicate the flanges on the original crossmember so I can weld the floor back down and support the vertical flanges to keep them from rolling under load. The bottom is thinner as I figured it wasn't the main support of the crossmember and I am pretty tight to the trans and wanted to keep the overall height of the crossmember fairly short.
The bottom was a complete shot in the dark. The vertical parts make sense at 3/16" but I probably could have made the bottom part thinner.
For what it's worth, the original crossmember looks to have been 14 ga (0.074")
Map63Vette
Well-Known Member
Been a while since I made mine, but I'm thinking I was probably somewhere in the 0.120" range on my material, mostly just based on what I could get in the local hardware stores at the time. Might even be more like 0.094 or something like that. I think I had the main profile pieces cut out of some spare sheet we had at work when it was on the table (and we were still allowed to do that). The main strap over the top and bottom I'm pretty sure was a hardware store special. Granted my car is also not really built for any kind of serious power, so I didn't go too crazy. Mostly just kind of looked at what was there and what was easy to get. Yours definitely looks beefier, so I doubt you'll have any problems.
I actually left my whole setup bolt-in. I thought about trying to weld the floor back down to it, but there was no way I was getting my rounded tunnel to lay any kind of flat on my upper crossmember piece with the tools I had at hand. I drilled a second set of holes in the t-bar crossmember further outboard to mimic the 4 bolt style of the original. I don't recall if I welded the crush tubes back in or just left them loose though.
racerjoe
Well-Known Member
are you having send cut send make the pieces?
DionR
Well-Known Member
DionR
Well-Known Member
I have to say I like SendCutSend.
I had a problem with the mount where the slots were too close to the bend line and they pushed pause on the order and then allowed me to make it wider so the part could be formed without making it useless. In addition, while I was fixing that I noticed that the front plate on the lower crossmember was missing a hole and the slot was just a hole on the DS end. So I uploaded a new file with that fixed and they swapped it out at the same time.
Now, I am sure if it had hit production with the hole issue before I saw it, I couldn't have gotten it fixed. Some of it was luck, but the fact that they caught the error and worked with me made it possible for both issues to be fixed. Very Thankful.
The issues with the holes is something that anyone using Fusion should be aware of. I like Fusion in that you can create a sketch and then extrude the sketch to create your part and if you need to change something you can update the sketch and the 3D part will be updated at the same time. You kind of "go back in time" when you make the changes so anything related to the sketch is also updated. The problem is that it doesn't always get things right when it updates.
I have a single sketch for my lower crossmember and the mount was about 3/16" too far to the PS side and about 1/16" low. Doesn't mean much since the actual part will get tacked to together in the car, but I wanted to to be accurate.
But when I moved the mount and it updated, it didn't do it right. Really easy to fix, but only if you notice it which I didn't. Usually it had issues with the full area and had weird triangles and such missing or added, but this time it dropped the hole and slot. I was checking the perimeter but didn't check the holes closely enough. Sure glad I had the opportunity to fix it.
I probably should have kept the mount sketch separate, that would have potentially made it more robust. And I need to learn how to use the sheetmetal functions, that might have alerted me to the issues with my slots. But still learning and feeling pretty good about where I am at right now.
I had a problem with the mount where the slots were too close to the bend line and they pushed pause on the order and then allowed me to make it wider so the part could be formed without making it useless. In addition, while I was fixing that I noticed that the front plate on the lower crossmember was missing a hole and the slot was just a hole on the DS end. So I uploaded a new file with that fixed and they swapped it out at the same time.
Now, I am sure if it had hit production with the hole issue before I saw it, I couldn't have gotten it fixed. Some of it was luck, but the fact that they caught the error and worked with me made it possible for both issues to be fixed. Very Thankful.
The issues with the holes is something that anyone using Fusion should be aware of. I like Fusion in that you can create a sketch and then extrude the sketch to create your part and if you need to change something you can update the sketch and the 3D part will be updated at the same time. You kind of "go back in time" when you make the changes so anything related to the sketch is also updated. The problem is that it doesn't always get things right when it updates.
I have a single sketch for my lower crossmember and the mount was about 3/16" too far to the PS side and about 1/16" low. Doesn't mean much since the actual part will get tacked to together in the car, but I wanted to to be accurate.
But when I moved the mount and it updated, it didn't do it right. Really easy to fix, but only if you notice it which I didn't. Usually it had issues with the full area and had weird triangles and such missing or added, but this time it dropped the hole and slot. I was checking the perimeter but didn't check the holes closely enough. Sure glad I had the opportunity to fix it.
I probably should have kept the mount sketch separate, that would have potentially made it more robust. And I need to learn how to use the sheetmetal functions, that might have alerted me to the issues with my slots. But still learning and feeling pretty good about where I am at right now.
racerjoe
Well-Known Member
Nice catch on their part. I'm anxious to see how it looks out the box from them. I've never had them bend anything.
Map63Vette
Well-Known Member
How you build model features relative to others can really bite you as well when you update stuff. More of an issue when you're dealing with a model you didn't build yourself, but easy enough to happen. So as an example, say you place one hole, then place a second hole and reference its location off of the first one. Down the road you find out that you need to tweak the position of the first hole, but when you move it, it will move both because the second hole is based on the first. Sometimes this is a good thing, like moving an entire mounting pattern by moving a single hole. Other times it's a pain because you only wanted to change the one hole.
Best practice it to try to dimension stuff relative to the stuff that affects it most. So if you need that hole to be 1/2" from that edge no matter what because it won't bolt up without interfering otherwise, you should dimension that hole off that edge in the model. Drawings are a bit of a different story, but it still kind of applies. Better practice with drawings is typically to limit all references (aside from patterns) to two primary edges (technically three I guess if you have patterns on more than one face). Essentially you want one "master face" that you can measure off of to limit tolerance stack problems.
A lot of CAD software nowadays will also let you know when you're features are fully constrained by changing the colors of the lines. Changing radiuses is one of the things that tends to push a lot of other stuff downstream around. One test I like to do after I have a sketch dimensioned is to grab one of the lines and drag it to see if it will move. If it's fully constrained it shouldn't, or if it's missing some constraint the way it moves will help you determine what constraints you need to add to lock it down so other changes won't mess with it.
Best practice it to try to dimension stuff relative to the stuff that affects it most. So if you need that hole to be 1/2" from that edge no matter what because it won't bolt up without interfering otherwise, you should dimension that hole off that edge in the model. Drawings are a bit of a different story, but it still kind of applies. Better practice with drawings is typically to limit all references (aside from patterns) to two primary edges (technically three I guess if you have patterns on more than one face). Essentially you want one "master face" that you can measure off of to limit tolerance stack problems.
A lot of CAD software nowadays will also let you know when you're features are fully constrained by changing the colors of the lines. Changing radiuses is one of the things that tends to push a lot of other stuff downstream around. One test I like to do after I have a sketch dimensioned is to grab one of the lines and drag it to see if it will move. If it's fully constrained it shouldn't, or if it's missing some constraint the way it moves will help you determine what constraints you need to add to lock it down so other changes won't mess with it.
DionR
Well-Known Member
So glad they caught it. I am sure there is some logic that flags it, but they could have just as easily shrugged their shoulders at the issue and sent it through.
I don't think they have had bending as an option for very long. I am excited to see how it looks, too.
DionR
Well-Known Member
Exactly.
Coming from a 2D realm in AutoCAD, the constraints was the biggest learning curve and source of frustration. There is power in the constraints, but for a rookie like me it sometimes was the rope I hung myself with. But it's just a level of experience and I figured it out to a degree.
I had problems with things like making the base of the lower crossmember wider where the upper corner was getting moved rather than just changing the angle of the line and length of the radius. So I had to learn how to lock the top radius location and force it to change the angle of the line. Not horrible, and there is probably a better way to do it, but it's what I found that worked.
Map63Vette
Well-Known Member
Every CAD software is different, so you just have to play around and learn the quirks. I learned on ProE in college (which I think is something else now), then used Catia for 13 years at my first job. Just changed jobs several months ago and now get to play with Inventor. Then I have Solidworks at home (and used it a little in college for a senior design project). Sometimes I want to beat my head against the monitor when using Inventor at my new job, but other times I find myself thinking "wow, this would have been a really nice feature to have had in Catia", so there are pros and cons to all of them. I often say that CAD software is like programming languages. They all do the same thing, it's just about learning the syntax and shortcuts. ProE called assigning depth to a sketh "pad", while other software calls it "extrude". Inventor drives me nuts because if you want to cut something, you still use the "extrude" function, but you make it "negative". Weird choice in my eyes, but once you learn how it works it's easy enough to deal with.
DionR
Well-Known Member
I learned drafting in AutoCAD but had a summer internship with Boeing and was able to get a bunch of training in Catia. I thought I might try and leverage that into a job with Mopar (I was a Mopar guy even way back then) but never pursued it and never actually used Catia for anything other than in training. I was exclusively AutoCAD (both 2D and 3D) until 2008. In 2008 I started working with a developer customizing Tekla (used to be Xsteel) which is a shape driven 3D modeling software for large structures. The school I went to later added classes in Solidworks, but way after I was graduated and I have never touched it. And until I bought my 3D printer, I hadn't used Fusion360 either. Even with access to Tekla, almost everything I did for myself was still AutoCAD and 2D.
Tekla is great for what we using it for, but it is a great example of how some software can fit better than others. Tekla does hot rolled shapes, can cut and cope flanges/webs, does plate work, creates drawings and is great for BIM and has a very robust API. My understanding is that Solidworks is great for machined parts and can animate assemblies to verify motion. But Solidworks would suck for my job and I would hate to try and make Tekla model widgets.
Map63Vette
Well-Known Member
Yeah, Catia was supposedly really good for shape design, but was probably way overkill for the place I worked. We basically did "burn and bend" design from sheet and plate, rarely anything that would take any kind of stamping or other forming. I always have to chuckle with AutoCAD because I feel so left out. I pretty much never used AutoCAD, but everyone I've ever worked with seems to know it well (or at least anyone older than me at the jobs I've had). My original drafting class in college was mostly paper based to learn the techniques, we didn't actually use software much as I recall. AutoCAD drives me up the wall sometimes because all I want to do is X, but since I don't know all the shortcuts and commands I usually spend a few minutes looking through ribbons and menus trying to find the one button I want. Both of my jobs had/have legacy AutoCAD data though, so I do have to use it every so often. Moreso at my new place, though both of them used it pretty exclusively for wiring documentation, which it does fit fairly well for. It's just strange to me since I'm so used to having a model linked with a drawing, so when I need to change something I default to looking for a model to update first.
Random.. Saw this over the weekend, while search for something in a group. It's for tremac. Would be easy to fab without complicated bends. I assume they had to lay it back because the design itself, is taller than 1/2 plate that has been formed
Last edited:
DionR
Well-Known Member
I can't find the info now, but that is the E-Body crossmember ToddRon did up and Modern Driveline sold. It's pretty much the same as the crossmember for an A-Body that I know Modern Driveline used to sell. ToddRon licensed their crossmembers to Modern Driveliine to sell but guessing the interest hasn't kept them alive.
Pretty sure the top crossmember is tipped forward to clear the shifter as (I think) the T56 puts it inline with the stock crossmember.
Interestingly, Modern Driveline seems to have cut all access to even look at the A-Body crossmember they used to sell. Now the link I saved just goes to a DIY and blog page. The crossmember seemed to be perpetually not available so I am betting they finally just pulled them from the website. With my motor in a different spot I couldn't use one either way, but sucks for anyone else who planned a T56 swap as I think it was a pretty good option (minus the $699 price).
The images are still on their website for now:
I used this as a foundational idea for mine. But I wanted flanges to weld the sheetmetal back down too I wanted more of a tube design and I needed it to fit my application. I've never touched one of theirs, only seen images online so I didn't copy it other than the basic idea.
Map63Vette
Well-Known Member
Yeah, I think everyone pretty much comes up with the same design when it comes down to it. The thickness is basically determined by the existing crossmember and usually one bolt hole is retained from the original setup as well, so only so many different ways to go with things. Looks fairly similar to what I made myself as well, though certainly much cleaner, lol.
Yeah the E-body version of the ToddRon T56 crossmember has the upper crossmember at an angle because the location of the torsion bar crossmember in the E-bodies is further back (longer torsion bars) and that puts the crossmember directly in-line with the shifter for the T-56.
Here's the E-body crossmember on the T56 Magnum vs the A-body crossmember
It's a damn shame that Modern Driveline isn't marketing/producing those crossmembers better, really glad that I bought one for my Challenger when ToddRonn was still making them. It's not like they're super complicated to make, but it just takes all the guesswork and trial and error out of mocking up and making one yourself. Sure, they were expensive but the time and energy it would have saved me mocking up that T56 in my half a dozen times to build my own crossmember was more than worth it.
DionR
Well-Known Member
I agree, it might be even harder to come up with a design that was different.
Interesting. I assumed it leaned forward, not back. Thanks.
I could see a change coming on stuff like this. In the past, the second investment after the development costs was buying/fabricating the parts to stock them. It made sense if you had to get them cut, welded together and powder coated, and if you do them in house having them done so you can just ship them makes sense. But if the design is such that the parts could be shipped direct from the cutting company and let the customer do the welding/assembling then it becomes much more of a build on demand rather than stock up prior to the demand. There are drawbacks in that you are kind of at the mercy of the end user being a quality individual as someone that has trouble putting it together can blow back on you. Or potentially over running your supplier if several copies are sold all at once. But for a small market product like this it makes more sense to hold onto the files and wait to cut them when someone orders them rather than find a space on a shelf to stock them. So no initial investment for stock and no money spent on parts that aren't selling.
The other issue with a setup like that is there is a cost reduction for multiples. So buying 10 sets means you can probably significantly reduce the price. But if you don't have the funds to invest and there isn't any place to store them, it makes some sense.
The other place this method doesn't work is if there are jigs involved. Unless you could sell the jig with the kit.
racerjoe
Well-Known Member
I agree. How can these things not be made for every Tremec, and 8speed along with a template of where to cut the tunnel and the sheet metal for the tunnel. Perhaps mopar people are just too cheap and want to do it themselves.
Denny's is pretty good if the motor is forward.. which his K members are. He offered to redo some parts for me to try to make it work in my case, but I really didn't want to play the ship back and forth game... I just needed to get down to it and do it.
DionR
Well-Known Member
WooHoo!! Parts are here!
Everything seems to fit up really nice. Very pleased.
Everything seems to fit up really nice. Very pleased.
340sFastback
Well-Known Member
I run the ToddRon cross member with my T56. Last weekend I went to drain trans fluid and found out I had to drop cross member down to access the drain plug. Not a big deal since won't be changing trans fluid all the time. I am leaning toward leaving it as is. But might come back and cut access hole so I can get drain plug out with cross member in place. Just a heads up something to be aware of.
Which T56 is that?
Drain plug on my T56 magnum is well clear of the crossmember...
340sFastback
Well-Known Member
Mine has a boss there for the drain plug but its not drilled and tapped for a plug like yours. Mine is the GM close ratio T56 Magnum that Silver Sport sent with the swap kit.
Last edited:
DionR
Well-Known Member
Thank. I saw your post about that but didn't give it much thought. Appreciate you pointing it out.
I have a drain plug there as well, plus one in the tailhousing like 340sFastback.
Weird.
-
Similar threads
