The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support Morrison Fabrications
Please Support Rix Racing

2G Building a stronger 5 speed transmission

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

FAQGSX222

Proven Member
224
77
Jul 26, 2015
Louisville, Tennessee
What goes into building the stock transmission stronger; it’s a question I’m sure many people have asked but most haven’t had answered other than the standard “buy better parts” answer. Stronger shift forks, upgraded gear sets and other parts that are extremely expensive through aftermarket businesses. What about the DIY guys , like myself, that are interested in learning to build one themselves by sourcing the correct parts from reputable sources? The best place to start figuring out what the process is for me is to search and ask questions on here since I know I always either find what I’m looking for or you guys point me in the right direction. In the long run I’m not sure if it’d be cheaper to just buy a stage 2-3 trans from somewhere or a used(I usually don’t see many but they’re usually $$$) or build a similar stage trans myself. Figuring out which parts that need to be upgraded is part of the learning process which others have learned through trial and error which helps others. I’m hoping that those who’ve built or are in the process of building a better and stronger transmission will share what there’re learning or what they’ already learned here on Tuners. At my age I’ve found myself focused on my DSM which may or may not be a good thing.:hmm:
 
Pretty much the only option is to get a 90 HD gear set and send it to a company like liberty gears to get them shot peen, cryo treated, and things like that. Or build a dog box trans if you are making big power. Evo 3 gears are a factory shot peened option with better ratios but they are no longer available new. And if you do find a new set somehow they will be $$$. You can upgrade to a 4 Spyder diff and a stronger lsd. But that's about it. Even with these options my trans has been breaking a lot lately. Shredding gears, shattering diffs, etc. Until someone reverse engineers the gears and makes a stronger factory helical gear set, were kind of screwed. Hopefully that happens one day. Even stock parts are becoming hard to locate anymore and are getting expensive.
 
"Buy better parts" isn't really an option either since very few of the good OEM parts are available new anymore. Aftermarket gear kits were never really an option other than PPG & PAR dogboxes (even Liberty's faceplated gears used stock gears). If you want a totally new synchronized gear set, you'll need to buy the $5,000 PAR synchro gear set, which is the only option now. The "Super HD" gear set and Evo III collection are very difficult to piece together these days.

Keep in mind that 'staged' rebuilds are transmissions that use stock gear sets with minor improvements to improve longevity. Shot-peening, cryo treating, deburring, etc give fractional increases to improve strength, but really, you need everything you can get when you're building a transmission that was initially designed for less than 300 lb-ft of torque. These aren't things you can do in your back yard. Transmission rebuilders can't build a high-torque helical gear transmission from scratch anymore, so you'll need to have a good core to work with. You'll need to re-use the shafts, gears and hub/slider assemblies. If you need replacements here, you'll probably have to buy yet another core and hope for better luck.

A basic rebuild can be done with new synchros, bearings and all of the small parts. This could be done in your shop if you know what you're doing. There are recommended preloads for your power level and such that have been published before. If you are going to have 400+ lb-ft of torque, you'll need to have the gears and shafts treated/improved, which drives up the price, since you can't do that yourself anymore.

In the long run, expect fierce competition for rebuildable cores of the strongest transmissions (with the HD gear set, as mentioned). You'll need to look way into the future when you make your decision here, knowing that some parts just won't be available when the time comes to replace things that are broken. It will make you second-guess destructive driving habits and find ways to make your existing setup live. Long story short, buy the best transmission you can find right now and keep it alive as long as you can.
 
Last edited:
Also, if you want to know the major weaknesses of DSM transmissions, this is a bit of a cheat-sheet.

In these transmissions, everything can break. They are weak. The 1st and 2nd gear aren't as concerning for outright strength (unless you have Galant VR4 parts), neither is 5th. Most failures will occur in the 3/4 cluster and in the center differential. Replacements for these parts are understandably the hardest to find now.

The 93-99 DSM gear set will shift nice but has the weakest 3rd and 4th gear setup due to the large-diameter single synchro design. The 3/4 hub and slider was a thinner version that will grenade pretty reliably at or before 400 lb-ft of torque. The 3rd and 4th gears do not like back-to-back pulls and will shred teeth with too much torque and heat.

The Evo III gear set was the best setup all-around due to the best ratios from 1-5 and double synchros on 2, 3 and 4. The forks, rails and hubs/sliders were forged and broached units which is great for strength, and the gears and shafts were shot-peened for added durability. It had a thicker hub and slider that was by far the strongest of the bunch. The 3rd and 4th gears and intermediate shaft could withstand a good amount of power, but applying 500 lb-ft of torque or doing back-to-back highway pulls would eventually strip these gears too. Evo III 3rd & 4th gears are still available, and the intermediate shaft from a DSM can be used, but the Evo III hub and slider to make it work together is unobtanium. The Evo 1-3 5th gear is extremely sought after for the taller .617 ratio, but these sold out years ago. Similarly, the Evo 1-3 1st and 2nd gears with the input shaft are long gone, too.

The DSM "Super HD" gear set was probably the strongest for 3rd and 4th gears. These came from 1990ish DSMs and would stand up to the highest torque but didn't have the best shifting feel. The 3/4 hub and slider was decently strong and was readily available, but used inferior materials to the Evo III. Unfortunately, those hub/slider assemblies are no longer available either. Properly built and well cared for, these would probably stand up to 600 lb-ft for a while.

The 91-92 DSM (aka "HD") had some strong parts as well, but the gears were a tad weaker. Some people used modified versions of these hub/slider assemblies with Evo gears when the Evo III hub/slider disappeared. It's not a terrible option.

You can't just throw the HD gears into a 93-99 transmission, it will require a lot more to make it work together. Similarly, you can't just use Evo gears in a 93-99 transmission either, without the proper hub/slider (and possibly fork/rail). Jack's Transmissions does list some of that information on their website.

OEM gears are all helical, synchronized units.
PAR's syncronized kit uses straight cut gears which are noisy, but much stronger. It will still require some parts that may be very difficult to find.
Dogboxes use straight cut gears and are noisy, but also 'dog' engagement which is really rough and sucks to drive on the street.

You'll see a common pattern here - don't do back-to-back highway pulls with a high powered car! This fatigues the gears immensely. Similarily, don't roll onto the power from low RPM because the torque spike on many setups around 4,000 RPM can make the 3rd and 4th weaker with each pull. If you're going to accelerate in 3rd and 4th, start from higher RPM to avoid the 3,500 to 4,500 range. Even in my light Evo with a transmission cooler and built to the max with the best OEM parts, I won't be acting like it's indestructible.

Avoid torquey setups. Yes, the 2.3L stroker is very nice to drive, but when combined with a big turbo, that torque is your transmission's worst enemy. A responsive 2.0L can be just as pleasant.

Launching harshly harms everything, but the transmission will twist the output shaft to the transfer case and the 2-spider center differential can break, which is very dangerous to the rest of the transmission, too.

Clutch drag can obliterate a freshly rebuilt transmission, and poor choice of clutch can shock, break, and wear out transmission internals too.

More info here: https://www.dsmtuners.com/threads/jacks-trans-super-hd-gears-and-durability-s.493495/
 
Last edited:
That last sentence nailed it on the head! I just happened to be scrolling eBay like always and found a fresh “91-94 hd drag trans” sold directly from jacks transmissions with warranty and couldn’t resist, I was planning on rebuilding my 90 trans this winter via tre, tmz or white she’d speed. But when I saw the ad I jumped on it.
 
To me, building a "stronger" 5 speed can't be done with OEM parts. Sure, some gears from XYZ years hold up better than others, but all the OEM gears are weak when you start transferring excessive amounts of torque through them for extended periods of time. With OEM components, it's not a question of if, but when your box will go down.

If you want a box that's truly stronger, you're looking at an aftermarket cluster. PAR and PPG are probably the most common. Those gearsets are expensive, and pricey to build. They also have longevity issues of their own, so it's not all rainbows and unicorns even after you take the plunge. This is a big reason you see a lot of drag and high power builds going auto.

OEM DSM transmissions hold up pretty well for the amount of abuse they typically see, IMO. Depending on power/torque levels, they can last a very long time. Driving habits are also a big factor on longevity. Don't expect it to last coupled to your 500+ HP stroker build, though. I think a lot of people don't fully consider the ramifications of their build choices until they are 3 transmissions deep, wondering why their 800hp 2.3L destroys everything it touches.

Lastly, things like your turbo, tune, cams, and many other factors are also mechanisms which can be leveraged to ensure the trans on even a high power build lasts. Configuring your setup to ensure peak torque happens at a point where it's less likely to clean the teeth off third gear down low in the RPM band is just one example of a solid option to make your trans last longer.

Luckily for us, there are options, and they don't all require building a *bulletproof* gearbox. Just understand whatever power you plan to make, it's got to get to the ground. Sadly (or not, depending on how you look at it), the 4G63 is much more capable at making power than our transmissions are at transfering it to the pavement. Use that as a planning consideration when building, and you'll be fine.
 
To me, building a "stronger" 5 speed can't be done with OEM parts. Sure, some gears from XYZ years hold up better than others, but all the OEM gears are weak when you start transferring excessive amounts of torque through them for extended periods of time. With OEM components, it's not a question of if, but when your box will go down.

If you want a box that's truly stronger, you're looking at an aftermarket cluster. PAR and PPG are probably the most common. Those gearsets are expensive, and pricey to build. They also have longevity issues of their own, so it's not all rainbows and unicorns even after you take the plunge. This is a big reason you see a lot of drag and high power builds going auto.

OEM DSM transmissions hold up pretty well for the amount of abuse they typically see, IMO. Depending on power/torque levels, they can last a very long time. Driving habits are also a big factor on longevity. Don't expect it to last coupled to your 500+ HP stroker build, though. I think a lot of people don't fully consider the ramifications of their build choices until they are 3 transmissions deep, wondering why their 800hp 2.3L destroys everything it touches.

Lastly, things like your turbo, tune, cams, and many other factors are also mechanisms which can be leveraged to ensure the trans on even a high power build lasts. Configuring your setup to ensure peak torque happens at a point where it's less likely to clean the teeth off third gear down low in the RPM band is just one example of a solid option to make your trans last longer.

Luckily for us, there are options, and they don't all require building a *bulletproof* gearbox. Just understand whatever power you plan to make, it's got to get to the ground. Sadly (or not, depending on how you look at it), the 4G63 is much more capable at making power than our transmissions are at transfering it to the pavement. Use that as a planning consideration when building, and you'll be fine.

:applause::thumb:
 
Im not super knowledgable when it comes to transmissions, but whats stopping someone from designing a stronger helical gearset? like, reverse engineering the stock gears with a stronger metal or something? Im sure it would not be cheap, but sooner or later even used OEM parts are not going to be an option. People are going to start to hoard them instead of selling them and we will be out of options. Dont they have CNC machines now a days that you can put the original gear in a machine, scan it, and it will replicate it? Again, im sure its not that easy or someone would have done it, but I feel like there is going to be a market for them soon.
 
I would love to know what alloy the OEM parts are made from, and what kind of heat treating and surface treatments they might have had during manufacture. Knowing that, you could guess at how much there is to be gained just by using a stronger material, like some kind of tool steel, or whatever material is great for gears.

I didn't know until reading above, that the PAR gearsets were available as synchro straight cut.
That seems to me like a pretty cool thing because you could have your synchros, and get rid of the thrust loading problem on the bearings that you have with helical gears.
But how much stronger are the PAR gears than the OEM - does anybody know?
 
It would be easy to check on some spare parts... I have the ability to check chemical composition of the alloy, surface to core micro-hardness readings, even grain structure to estimate the heat treat schedule.

But I dont have any spare gears to use for destructive testing. Does anyone have some broken gears {edit: any and all trans parts} to send me? Then we could, as @We're on Boost mentioned, make some intelligent guesses about what could be gained in metallurgy and heat treat.

I also don't know if anyone else is developing trans parts right now. I wouldn't want to dump a bunch of time into the project just to find that someone is trying to get rich selling the same thing.
 
Are you a gear specialist? There are a lot of nuances that go into gear design that are not in the normal engineering curriculum. Also where are we going to get them made. It's easy enough to draw gears on any cad software, but getting them made is tough. From what I've seen the original gears were 4135. And they were possibly nitrided instead of traditional case hardening. Lots of jap companies were doing that during this time frame. Found a couple sae papers on it. Nothing from mitsubishi, but it wouldn't suprise me.

Anyway, if we are going to make new gears, why make oem shit. Don't waste time trying to make new oem like gears out of better metal. way to many shortcomings in the factory stuff. At the very least we need the stacks to be longer, for a little bit more gear width, and take the undercut out of 4th gear. Then we need a helical lsd center diff with a replaceable gear.

Personally i've only broke 3rd gear, and 4th once. 3rd just gets killed by fatigue, and 4th has the stupid undercut. 1/2 stay good. We could probably get by with a set that made 3rd the full width of the intermediate shaft, possibly a profile modification, along with fixing the undercut on 4th. That would be fine for 90% of users. but then we are boned on sliders. IMO the liberty stuff isn't an option. just get a ppg. The straight cut whine is small in comparison to the dogbox pita.

I had started to work on a set of cases that would put 3kgt 5 speed gears in a dsm. Would have been sweet, 1000ft*lbs capacity, a 45/55 power split center diff, with a 35/65torsen option, same ratios as a dsm trans with evo3 finals, I ran 9's on that exact setup... Ended up buying a ppg since I figured no one would spend the $3500+ it was going to take to make it profitable to do. Crying about ratios, and other stupid stuff.
 
It would be easy to check on some spare parts... I have the ability to check chemical composition of the alloy, surface to core micro-hardness readings, even grain structure to estimate the heat treat schedule.

That would be a really cool thing. Kind of amazing that you have access to stuff like that.
I just took a quick look at what steel alloys might be a "best" upgrade material for gears and shafts.
Looks to me like Ferrium C61 would be one great possibility.

https://www.questek.com/ferrium-c61.html

I literally only spent about 15 minutes looking.
But look at these numbers for Fracture Toughness. Units are Ksi-Sqrt(inch)
Higher is better:
C61 - 130
9310 - 85
300M - 50
4340 - 50

Although 300M has slightly higher Yield strength.
 
I wouldn't consider myself a specialist by any means. Have a designed gear sets in the past? Yes. Professionally? yes. For power transmission at this scale? NO.

Getting them made may not be as difficult as you might expect. I have a rolodex on my desk with dozens upon dozens of machinists. Two or three come to mind when I am picturing the gears/hubs/sliders.

It will not be cheap to prototype these parts, and at the volumes we would be looking at, I would expect the price point doesn't drop off much.

I also have experience with case hardening and specifically nitriding and carburizing.

My coworkers are all car guys, so I can probably get advice/guidance on metallurgy from one of our metallurgists as well. Our company is turning 100 next year and metallurgy is somewhat of a point-of-pride for us.

As bastard said, there are many nuances to gear design. I actually like the sound of his project much much better...

Bastard, when u made the call to can that 3kgt setup, were transmissions this hard to find? $3500 sounds amazing, depending on the specifics!
 
Last edited by a moderator:
There are very limited new 3kgt parts. But the 5 speeds are cheap.

As got materials, there's a lot more to it then just fracture toughness and yield strength you got to be able to machine it and that ferrium and 9310 can be a cocksucker.


If you guys really want this to be successful it needs to be treated like a real engineering project and mostly done behind closed doors otherwise you'll have every Tom Dick and a****** chiming in with non sense. You're going to need to put together a list of all the problems you want to solve. And you have to look at the whole system the engine the turbo the clutch the rear end. Everything affects everything.

A good place to start is make a 2g 3/4 hub and slider that will survive.
 
Pretty much the only effective option to handle real power anymore, with parts availability, is a PPG gearset, Boostin billet bearing plate and 300m output shaft, a Quaife or KAAZ LSD front differential, and call it a day.

9310 for the transmission gears is plenty. For the transfer case, I was going to have them made from Ferrium C61 with 300m or 4340 for the output shaft.

Heres nice reading material on Ferrium materials:
http://gearsolutions.com/features/processing-new-gear-steels/

To start from scratch, and batch volume, you are easily above $50k cost on 25 units for just the transfer case gearsets , AT COST. 100% downpayment, wait 6-9 months, and then you have to sell them at a price to make it worth your while to fork over that kind of $$$$. I've already done this legwork in the past.

For the transmission, it is really useless to reinvent the wheel for making another entire helical gearset. I do know a place that could make the gears , but it again, is not cheap and is overseas with no material data disclosed, or real world testing at high power levels to get a solid durability database.

The biggest challenge is determining the most effective individual component to manufacture. My feeling is make a stronger 90-92.5 style 3/4 hub and slider is the best compromise.
 
Phht. 3,500 bucks. I spent 3,900 on a road race trans years ago from Jack's. I think if the price was between a built trans and a dog box more people might hop on. A staged rebuild can handle roughly 400 to 450 ft lbs. Then a dog box , is more than capable, but 8 grand is out of reach for many, and that's more than thier entire build.
 
Phht. 3,500 bucks. I spent 3,900 on a road race trans years ago from Jack's. I think if the price was between a built trans and a dog box more people might hop on. A staged rebuild can handle roughly 400 to 450 ft lbs. Then a dog box , is more than capable, but 8 grand is out of reach for many, and that's more than thier entire build.

If there was a helical gear set made of stronger metal in-between the price range of the two you can count me in. I'm sick of spending 2k on a trans that doesn't hold 400tq. I've shredded 4th and now 2nd gear making around that tq range. It always wipes out a bunch of other crap in the process of letting go at 6k+. A stronger 3/4 hub and slider, upgraded diff, and upgraded gear set would make me much more comfortable with spending the money to actually rebuild a trans instead of just buying used ones. I personally don't want to spend the money on straight cut gears because I don't want the annoying whine. Mine is mostly a street car.
 
How many people are trashing hub/sliders? and HOW? I am put my synchro trans back together to run for a while, and I'm still on the original hub/sliders that I bought back in 2009 when I first "built" a trans.

Anyway, the whine of the ppg is nothing at all imo. In 5th gear cruising it's pretty minimal.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top