Inside...700R4 and4L60E TorqueConvertersby Doc Frohmader

Early on a lot of us knew the700R4 as an ‘overdrive TH-350’.In terms of size and capacity theyaren’t all that far apart, but themysteries and agonies of theoverdrive and lockup took sometime to thrash through. I can stillrecall when local transmissionshops would either refuse to workon them at all or refuse to warrantythem. In the first year of produc-tion at least 43 upgrades weremade to just the valve body. Theywere frying like dropped eggs on asteel street in august.

There were lots of individualreasons why they had trouble.Some had to do with a new tech-nology and failures not anticipated by theengineers, some had to do with the fact thatmost people who bought them had no cluehow to drive them, and some had to do withbeing a bit too light or having componentsmade of the wrong materials. Of all theproblems, though, one was a consistent andpersistent problem that remains to somedegree to this day.

We still have problems with convert-ers. They are found unbalanced, out ofround, with too much hub run-out, poor ordamaged clutch materials and/or damagedcontact surfaces, bad bushings and/orbearings, having incorrect internal clear-ances, loose turbine fins, and incorrect totalheight - and that’s just the ‘new’ or ‘rebuilt’ones. Actually some ‘rebuilts’ are no morethan washed out and repainted used stuff.Others are opened and the worst of the bad

parts swapped for others and welded backtogether. Some get new parts but they areof poor quality. A vast majority are not wellbalanced or are out of spec. When convert-ers fail, they throw debris throughout thetrans and pretty well thrash the works.

Price alone won’t help you either forguarantees. I’ve seen 50-dollar parts-houseconverters that worked well and high dollarhigh performance units turn out to be gar-bage. On the other hand, there will alwaysbe a ‘no free lunch’ clause to deal with.What I mean is you can’t reasonably expecta throw-down converter for el-cheapoprices. The best protection is to know whoyou are dealing with and know what to lookfor. If you can’t be sure what you are get-ting, maybe it’s best to move along to an-other source.

����������������

The stamped numbersin many converters tellyou details about stallspeed and impellers.See sidebar #1

The code tags havea four-digit lettercode whichprovides moredetailedinformation. Seesidebar #2

The converter is cutapart in a lathe whichcuts away the weldholding the halvestogether.

You see, the failure of a converterremains one of the most common failures ofthe 700R4/4L60E. It still has the samedevastating result on the rest of the trans.The cost of such failures often run well overa grand and if you compare that to a qualityconverter, it’s well worth whatever the con-verter costs. Hey, the trans is only as goodas the weakest part!

That’s the place I was this last springwhen I was fighting still again with a vibra-tion problem I was convinced had its originsin the transmission. The unit had beenrebuilt in front of me and I was as certain asI could be it was right. Still, my old 700R4had a cyclic vibration driving me up a wall. Ihad no confidence in the truck, was sure itwould sooner or later fail and leave mestranded, and I could feel my wallet heatingup from overuse. I mentioned this to GregDucato at Phoenix Transmission whileworking on another project and he told meto drag my butt into the shop with the con-verter and he’d school me. Not one to turndown a challenge to my limited brain power,I did just that. The resulting experiencemade me realize the difference between thejunk and the primo stuff and I had to shareit.

Good, Bad, and UglyThe converter that is not completely

disassembled is a mass of uncertainties. Idon’t care how good you think you are atjiggling and shaking and sniffing and prob-ing, there’s too much inside that can gowrong. For that reason Greg tears everyone down to components. Granted, heprefers using ‘virgin’ cores - those notscrewed up by the amateurs - but you cantell a lot about what is inside by two basicitems. First, there is a stamped numberaround the housing which corresponds tothe stall speed or fin angle (see sidebar).

William has just the right touch to pop the converterapart without beating it up. The hammer splits anyremaining weld material.

The front edge of each half is trimmed so it is cleanand even and will allow the parts to fit precisely.

The front half is alsoturned to clean welddebris and deepenthe shelf cut so thehalves can beassembled at thecorrect height later.

A damaged hub.Rather than try torepair or re-cut thehub, Phoenix willremove it andreplace it with a newhardened hub.

The hub mountsurface must beturned flat andsmooth or it couldend up defective.

The front half isflipped over and themounting lugs aretrued. You’d besurprised to find howmany of these arecrooked, beat up, orstripped.

The turbine on theright is an early multi-part unit with spot-welded hub. The oneon the left is the late-style one-piece withfull-perimeter weldedhub that is thepreferred part.

The clutch rides onthe front half of theconverter cover. Thecontact area must besmooth, clean, flat,not burned, and notgrooved or it getschucked out. There isno effective andeconomical repair fordamaged parts.

Second, many will still have the bar-coded GM sticker near the hub which givesvery specific information if you know how toread it (see sidebar). Greg tries to start withthe most appropriate cores for the specificapplication.

After draining the converter, it goes tothe lathe where the original weld is cutthrough. This allows the front and rearhalves to be separated and turned to pre-pare them to go together again in a nicetight slip fit. As happens, the core we builtalso had a worn input hub so the rear hous-ing was chucked up, the old hub cut off, andthe housing turned flat. The bad hub will bereplaced with a hardened hub and fullywelded again.

The other half also gets attention.The threaded lugs are checked for damage,re-welded as needed, and then the housingis chucked up in the lathe jig to true thelugs. This makes sure all are in the sameplane and at right angles to the centerline.It’s a common source of run-out in anyconverter.

If the clutch contact surface in thefront half is in good repair or it can becleaned up with light abrasive, the housinginspection and prep are nearly done. If theclutch surface is in bad shape, Greg dis-cards it. To cut it down could weaken it.

The remainder of the internal partsare thoroughly cleaned and all are in-spected. There’s a lot to learn about thecondition of the converter and where poten-tial problems could lie by taking a carefullook at the parts before any are discarded.

Sub AssembliesOne of the first steps is to replace the

hub. I can understand why some of the fastand dirty builders don’t do this. It takes agood piece of equipment operated by some-one who knows something about machiningand measurement.

The new hardened hub sets on the converter housingin a special jig. The housing rests flat and the conicalfitting locates the hub in the dead center.

Minute adjustments are done according to a dialindicator. The hub must be indicated to have no morethan .001 inch of run-out.

The entire jig is part of awelding rig. Theturntable rotates and aperfect, consistent weldis made around the hub.This robot welding is farbetter than can be hand-done.

It’s not enough to know thehub was within spec beforethe weld. It is re-checkedafterward to make surenothing was moved ordistorted. Attention to suchdetail is what makes a greatconverter.

On the right is the wovencarbon fiber clutch used in1998 and later severe dutyunits. On the left is thecarbon composite clutchwhich is a durable and heat-resistant material used inmost applications 1992-98.

�

� �

������

���

�

������

On the left is apaper faced clutchwhich is only goodfor the lightest-dutyapplications andwhich Phoenixnever uses in itsconverters. On theright is the Kevlarmaterial Gregbonds to upgrade orreplace wornfacings.

The common clutchdamper units. (A) isthe early HD damper,(B) is the late HD andperformance damper,and (C) is thestandard car and lighttruck unit.

Blade angle is the waystall speed isdetermined. The lowstall impeller fins angleto the right for lessaggressive fluidmovement. The highstall unit on the righthas blades angled tothe left which reallyincreases fluidpressure against thestator.

The housing is located in one side ofthe machine jig where a conical supportpositions it. On the other side a secondconical fixture locates the hub so it is exactlyon the centerline when the cones arepressed together. From there, WilliamBuxton uses a dial indicator setup to checkand correct run-out on the hub. When run-out is determined to be no more than .001,the hub is welded on with an automaticwelder to ensure a clean, precise, andevenly-heated weld all around the hub.

Next Greg determines which clutchand damper unit he’ll use. There are fourtypes of clutches to choose from. First isthe paper clutch. This is the low-end ofperformance and used only in low-perfor-mance applications and light vehicles.Second is the carbon composite clutch.This is a good quality material used by GMand it stands up well to most uses. Third isthe Kevlar clutch which is a high-heat mate-rial and a common replacement material forhigher performance converters. It was notused in OEM units. The toughest is thecarbon woven clutch material used onextreme duty 1998 and later units. It wasused especially where the trans was ex-pected to cycle often between lock-up andnon-lockup conditions. In many cases it’san over-kill.

The clutch dampers come in threeflavors. First is the early HD performanceunit which is found in high performance andsome diesels prior to 1992. The second isthe late HD performance unit used in nearlyall 1992 and later 700R4 and 4L60E’s. Thelast is the standard car and light truck unitused in most 1980-91 GM lockups like the350,200, etc. These parts are designed tocushion the application of the clutch muchthe same as a dry clutch plate works. Forthis application we used the late style HDbecause it is intended for truck use andtrailer towing.

By the way, the clutch material isrenewed by removing and cleaning off theold material and replacing the clutch mate-rial with a pre-cut section of the appropriatematerial. The new clutch is positioned onthe clutch damper so it runs true and in-serted in Greg’s ‘easy-bake’ oven. Actuallythis special oven cures and bonds the clutchmaterial to the damper at about 400 degreeand under pressure.

Next up is the impeller. Impellershave fins which are critical for a couple ofreasons. The impeller and stator are con-nected by fluid to transfer energy betweenengine and trans so the angle and directionof the impeller fins affects the stall speed.The more aggressively the fins angle intothe fluid flow, the lower the stall speed. It’sdesigned-in slippage under power. Asnoted before, there are several differentpitches from GM to choose from and thenthere’s modified fins. Greg can alter the finangle to develop just about any stall speedyou need.

While stock-type converters or low-performance units are less likely to have aproblem with fins loosening or breaking offunder hard use, Greg spot welds each finusing a silicon-bronze and a TIG welder.This insures the fins won’t go loose and failwhen you put the hammer down on ahealthy engine or under heavy load.Welded fins also reduce flexing under hardloading so there is less likelihood of metalfatigue, vibration, or erratic function.

In addition to altering blade angle as required, Williamdeftly uses a punch and hammer to tighten every finprior to welding. More important small details.

Each and every fin isspot-welded with a TIGmachine by hand. Gregsays it’s important notonly to weld each one,but make sure eachweld is exactly the sameand holds well. It takesskill and practice.

Yep, both sides!Greg is careful tomake the spots verysmall and staggerthe welds to keepheat soaking down.Productionconverters don’t getthis because it takestoo much time.

No, Greg isn’t makinglunch - he’s setting aclutch in the bondingoven. The Kevlarclutch material isglued to the plate,making sure it iscentered and in itgoes.

SIDEBAR #1 - converter body stamped codesMost converters will have a number or

letter stamped regularly around the outside ofthe housing. You can get a good idea of whatyou have from these stamped codes. Althoughthe stall speeds are quoted, these vary depend-ing on the performance level of the engine,vehicle weight, and gearing. It’s good for com-parisons.

#4 - 1211 rpmdiesel engines

#5 - 1611 or 2025 rpmtwo different stators were used which changedstall

#6 - 1397 rpmtypical 5.0L and 5.7L unit

#7 - 1654 rpmused in V-6, 5.0L, and 5.7L

#C - 2075 rpma #7 with brazed fins and high-stall statorused in 1996 4.3L jimmy, Blazer, Bravada

#H - 1397 rpma #6 with brazed finsused in 4.3L, 5.0L, 5.7L

#K - 1211 rpma #4 with brazed finsused in diesels

#L - 1654 rpma #7 with brazed fins

Sidebar #2 - Bar code or tag decodingGM used bar-codes and tags on many

torque converters. A lot of detailed info is onthem if you know how to read them. The firstdigit of the four is application. The second isstall speed and impeller type (brazed or non-brazed). The third is the type of clutch anddampers used. The last is mounting lug type.First Digit:B=250C and 350C trans.C=200C, 2004R, pre-1985 700R4, and 325-4LtransD=1985 and later 700R4 trans

Second Digit:B=2025 rpmC=2075 rpm, brazed impellerE=1654 rpmF=1611 rpmG=1397 rpmH=1397 rpm, brazed impellerK=1211 rpm, brazed impellerL=1654 rpm, brazed impeller

Third Digit:These are confusing to most people and

relate to internal valving and capacities thatwon’t do you much good without some detailedtech experience. On the other hand, if the digitis a G, H, or L the unit has the high-end wovencarbon clutch.

Fourth Digit:C= 3 round lug gas engineD= 3 round lug diesel engineE= 6-lug gas or diesel engineF= 3 square lug gas or diesel late styleG= 3 square lug gas or diesel late style

High pressure and400 degrees makethe bond permanentand help the clutch hold upunder high-torqueloads.

StatorsStators make a big difference in how

a converter functions in terms of stall speedfor example, but I found there’s more to it.First off, the early units used a smallerbearing which had a separate staked-inrace. The preferred stator used a large one-piece enclosed Torrington-type bearingwhich is much stronger and less likely to fail.This is an optional upgrade Phoenix uses onall 300+ hp applications but one that mightbe a good idea if you are putting it behind ahigh-performance engine, a heavy vehicle,plan to do a lot of towing, or foresee a lot ofmiles on it. It’s another one of those optionsthat depend on comparing the cost to the

benefit and you have to be the judge of that.Next there’s the high or low-stall

aspect. In truth, there are only two basicstyles according to Greg. The first is usedin almost everything and is the standardlow-stall unit. The second is somewhat rareand was used in a few 4.3 liter S-10’s andturbo’ed cars like the Grand National Buicksas a factory special. Contrary to myth,these were never used in Camaro or Cor-vette units. The High-stall unit has fewerblades and therefore allows more drivingfluid to bypass the fins and offers less areato be pushed against.

The stator on the Rightis the rare high-stallpart used in a very few4.3 liter S-10’s andturbo cars but never inthe Corvette orCamaros

The stator on theright is the late-style part with thelarge one-pieceenclosedTorringtonbearing. Theother is the earlystyle with thesmall Torringtonwith separatestaked-in race.The right choiceis obvious.

For the most part, the stall speed isdetermined by the pitch of the turbine inthese converters. As you may recall fromlast month, there are some serious differ-ences and Greg can also tailor the turbinepitch to change stall speeds to suit yourspecific application.

The large bearing(opened up for thepicture) not onlyhas more bearingrollers but morecontact area tospread thecontact energyacross. Lessfriction and lessheat buildup.

The Sprag cage on the bottomshows the kind of extreme wearthat can cause trouble. The cagealigns both rollers and springs forcorrect function.

The cage sets in the bottomof the stator after the stator ischecked for damage orcracks.

The Sprag cam sets over the top of the cage to retainit. The cam must be in perfect shape or it will rapidlyfail.

The Sprag innerrace is next. Itmust seat in thestator fully andwithout play. It is ahardened part thattakes a lot of stressso it is alwaysclosely inspectedand replaced foreven minor wear.

You can see how therollers fit into thecam. They requirecorrect clearances soany wear is a deathwarrant.

Inside the stator hub aSprag unit is located. ASprag is a type of one-wayclutch and takes a lot ofabuse. It keeps the enginefrom free-wheeling whenyou lift the throttle andhelps multiply torque onacceleration. At the bottomis a Sprag roller cage whichkeeps the rollers andsprings in alignment andprovides a place for therollers to rub that’s tougherthan the aluminum stator.This is one of those partsthat always gets thoroughlyinspected and replaced ifthe y show wear. The samegoes for the Sprag innerrace, and the rollers andsprings (most are replaced).These can make a bigdifference in how long theSprag works right, and thecost of the parts is not allthat high. It surprises methat some builders will

forego a couple bucks in new parts, makingminimal extra profit but a lot more failures.Unfortunately the whole transmission indus-try is infected with too many of these hacks.

Not all components require upgrade,though. For example, GM units use aphenolic washer on the back of the stator

and it works fine in manyapplications. Phoenixprefers a Torrington bearingand spacer conversion intheir performance units.The conversion requires analuminum plate machinedto the stator and the newbearing which rides againstthe turbine. For seriousperformance this is a realasset, but for the vastmajority of streetable ve-hicles the phenolic washerwill last and perform per-fectly.

The accordion-like Sprag springs keep the rollerspressed to one side until the rollers and cam areengaged and return the rollers. Broken, damaged, orcompressed springs won’t do.

A phenolic spacer is used in stockstators and work well in mostcases.

An aluminum spacer is installedinstead of the phenolic spacer forhigh performance or severe use.

The spacer is staked intothe back of the stator tokeep it from moving in andout.

If you want to eliminate thelock-up, the entire clutchassembly can be replacedwith this machined hub/spacer.

This phenolic insertsets end-play andseals the input shaft tothe clutch hub as wellas determines overallconverter height.They come in varioussizes for adjustments.

The late HD clutchdamper with the Kevlarclutch face seat in thefront half of theconverter housing.

The modified andsilicon-bronze weldedturbine is next in lineover the clutch.

Speaking of conversions, there is aspecial hub made to replace the entire lock-up clutch assembly to eliminate lock-up.This is sometimes used in off-road or veryhigh performance units where the clutch iseither not needed or has the potential tocycle too often or even outright fail do toextreme abuse. In almost every case, astreet vehicle performs best with a properlyfunctioning locking converter.

AssemblyOnce the sub-assemblies are rebuilt

and/or modified, it’s time to assemble theconverter. One of the first items of concern

is a phenolic spacer that fits between theclutch damper and front housing. These aremade in selective sizes and are used to setend-play, overall height of the converter unitand also makes the seal between the con-verter clutch piston and input shaft. Justslapping a new O-ring on it won’t make itright. If worn they are junk and may leakaround the shaft even with new shaft seals.

A lot rides on the converter assemblyhaving the right internal end-play and havingthe total assembled height of the convertercorrect. No matter how good the parts, ifthe specs are not made right you’ve still gota load of junk.

This Torringtonbearing is theother half of thephenolic spacerreplacement. Itgoes between theturbine and thestator.

This is the completeinternal assemblywith the stator on top- ready to cover.

For this reason, Greg determinesand/or sets the height of the internal compo-nents. They have to be within a narrowrange to function correctly. This overallheight is compared to the to the height ofthe converter housing to determine the totalinternal end-play and the final height of theconverter. I’ve stood and watched as an-other large builder slapped converterstogether without once checking this, so I cantell you from experience not all are alike.

On top of the clutch the turbine unitwith the silicon-bronze welded fins, theTorrington bearing for the HD conversionfrom the phenolic spacer and the as-sembled stator/Sprag unit are installed.Finally the back half of the converter isslipped into the front half.

From here the converter assemblygoes to the welding machine. The machineis also a jig which allows Greg to positionthe two halves precisely where they mustbe. The internal over-all end play is set bythe total height of the converter housing, sothe halves slide in and out to get to thispoint. You may recall from the first part howWilliam Buxton turned the housing halvesso they had a slip-fit so this operation couldbe completed with concentric housings.Both the height and the run-out is checkedand set with a dial indicator before anywelding.

Welding starts with a series of alter-nating tacks which fix the position of thehalves and make concentricity easier toretain. When this is done and the specschecked, the entire circumference is ma-chine welded for a consistent weld. Weldsare critical for several reasons. First issimple penetration and strength. Second isthe need for air-tight welds to prevent leaksor seepage. Finally, irregular welds makebalancing harder. If you see welds that lookpoor, they probably are and you should lookfor another source.

The height and run-out are checked I the weldingmachine/jig before any welds are made. Alignment iscritical and can make or break a converter.

You don’t back-yard a converter. It takes expensiveand specialized equipment to get it right, like thiswelder/turntable/assembly jig made just forconverters.

Final TestsOnce the welds are done, the con-

verter is double-checked for run-out andheight. After the converter has cooled, itgets pressure tested for leaks at 120 psi.For obvious reasons, even the smallest leakis unacceptable. If a minor pin-hole is foundit can be spot-repaired, but if there is morethere was a problem with the weld thatrequires re-opening and re-welding theconverter. A visual inspection to seek outany defects or flaws in done and the mount-ing lug threads are chased.

All Phoenix converters are balancedto zero on a special converter balancer.This is another area where the cheap con-verters fail. In the last few months I havehad some checked and found 20-30 gramsof imbalance to be common. The result isincreased wear and vibration problems.Greg sees it as not only an assurance theunit will work smoothly and last a long time,but an indication of potential problems. If aconverter shows up a large amount out ofbalance, there is some concern that one ofthe components is defective.

I installed the converter used todemonstrate for this article in my own truckbehind a tuned-port small-block and have tosay it works perfectly and is completelysmooth and vibration-free. The converter Iremoved was a parts-house item. Ichanged only the converter and eliminatedan irritating vibration. The converter I re-moved and took to Greg was checked outand found to be in such poor condition thatrather than rebuild it it was tossed in thescrap. It’s nice to be able to identify aproblem and know it is resolved.

Clean, strong, andvery consistentwelds are made inthe machine welder.Bad welds meanleaks and imbalanceproblems.

All converters are pressure-tested at 120 psi. Noteven the smallest leak is acceptable.

Even details like the threads in themount lugs are checked and cleaned.Shops that do this kind of quality rarelysee failed returns because they payattention to details. How many timesdo you want to replace a converter?

That zero on the balancerreadout means just what itsays. Zero balance is not anoption if you want a smoothand durable converter.