BASIC  TRANSMISSION

by
Brent Covey

 Wes Cauglin originally told me about the extra disk. I was then told the same thing last summer by a guy who owned a transmission repair shop in NM (and towed my disabled GMC).

There are two versions of the clutch drums that use a different number of plates. I am not certain all GMC's got the extra plate, but would imagine all 455 versions did at least. There were significant changes in the 1977-'78 car versions of this transmission and they may have also been changed in the GMC. The extra capacity clutches can be added to units that do not contain them, they're interchangable. Most THM 400 performance parts can also be used in the THM 425, although they are usually not useful for us.

How about differences between the Eldo and Toro trans?

Theres differences between the Eldorado, Toronado, Toronado GT and GMC as
well as year to year changes. Most relate to speedo gears and shift scheduling, and 1966-'67 versions all came with the switch pitch as well. 1975-'78 cars use a very low stall torque converter, something like 1700 rpm. It is likely the early transmissions do not have the extra plates, and likely that the seventies cars do. The extra plate(s) is a way GM could add shift quality without hurting durability, the essence is more plates lets
you shift at lower pressures without losing torque capacity. You retain good torque capacity from the extra surface area, and can leave the pressures low for imperceptible engagement. Theres also two oil pump sizes used in the THM 425 and some other little things that vary year to year.

All that aside, any of the THM 425's would be reliable in a GMC as they all even in the 'weakest' versions have a great deal of extra power handing capacity. Something to keep in mind also is the biggest loads the transaxle sees arent at full throttle, but at closed throttle coasting in low or second range. The engine makes 350 ft/lbs torque but the 12,000 lbs coach feeding power back into the unit can hit it with around ~6,000 ft/lbs
momentarily on a forced downshift from the tires.

Its always worthwhile to try to retain your original transaxle during a rebuild in any event, GMC had the best cores used by GM and the GMC specific shift scheduling and valve body should be retained if possible.
--------------------------------

How does the "shift Pack" affect the THM-425?  If the transmission is tuned
 for standard shift points wouldn't the higher shift points on the "package"
 put more strain on the transmission, especially on deceleration when going
 down hill?  Last point, do you recommend the "shift package" or not?

Generally shift kits change the transmission shift schedule, and increase the pressure relief setting for mainline pressure regulation.

The way they sell them is by how different they feel after they're installed, the transmission has 'positive' (rough) upshifts at different speeds than stock and you really notice the difference- kind of like brand X shocks for your car are so stiff after a change you are sure the originals were worn out and dont doubt its an improvement.

As I alluded to in the post addressing Patrick's questions you really dont want to lug the transmission below the converter stall speed. Doing this just heats up the oil and operates the engine under heavy load at speeds its not making a lot of power anyhow. If you are under load, the transmission should automatically select a lower gear, or remain in one until the load has passed. Most shift kits dont. If the kit doesn't kick down at part
throttle to provide response, you must manually select a lower gear and do it yourself.

The higher pressure regulators arent a good thing either, really. The main power drain in the transmission is the pump, which is of fixed capacity on a GMC and draws tremendous HP at high speeds or pressures. The object is to run it at the LOWEST pressure/flow that will still keep the clutches applied and not slipping. If the new pressure regulator spring adds 1/3 to transmission mainline pressures, all you're doing is making the transmission use 125% the fuel it did with the original one. The original spring is more than ample to keep clutches from slipping, so theres no benefit to the high pressure spring. Also, the mainline pressure isnt the MODULATED pressure
which is a proportional to engine load circuit that does most of the actual work, it operates at a lower pressure than MAINLINE- Modulated pressures are controlled in Drive range by the vacuum modulator which varies the pressure to keep the shifting at optimum pressure for smooth engagement.

The impression of a nice 'positive' shift is just that, an impression. The bumpiness of an upshift is not any indication of the wear and tear taking place in the transmission really. All the shifting going on in there occurs not by moving gears but by applying 'brakes' to stop or lock together various rotating parts. The engine makes varying output with speed as well. The object to get a smooth shift is to switch the braked parts off at the point the output shaft power remains the same amount of torque in the new ratio. Power is  anded off from one brake to another with little relative motion and if timed perfectly is  perceptible. If they're timed poorly, you feel them lurch into position.

In the GMC, the original shift scheduling is a little too passenger car to be perfect, it shifts up much too early at part throttle and much too late at full throttle. Some of the better kits address this partially by holding  lower gears at part throttle longer to keep the converter out of the loop as much as possible, which is a good thing. The lower full throttle upshift speeds also will be some benefit for general performance, if they arent too low.  They all leave much to be desired in whats called 'detent touch' downshifting, which is driving about 25-40 mph and tickling the gas pedal for a downshift to second gear for better response, rather than having to mash it and force a downshift. The original GMC shift scheduling is better than aftermarket in this respect, but none of them are that hot, most THM 400/425 cars are the pits as well about this aspect.

To specifically answer your questions, the shift kits higher part throttle shift points would be advantageous and not injurous to the transmission as they are only higher at part throttle and will help keep it cool by keeping rpm up and letting the converter stay above stall speed most of the time. The lower speed full throttle upshifts compared to stock may reduce wear and tear also, although it would be minimal benefit. Part Throttle 3-2 and 2-1 downshifts are poor in every case, but better with the stock setup. Manually
selected 2 and 1 ranges operate a mainline booster circuit that ups the pressures substantially to ensure theres no slip, and there is an extra band engaged to provide braking.

I feel the factory valve body is best left in there, and dont feel the shift kits are worthwhile. There are small modifications that can be made to the original valve body to improve the 3-2 part throttle downshift by substiution of parts from six cylinder or other low performance THM 400 units that had better detent touch downshifting, and small improvements can be made to perk up city traffic and part throttle response this way. Using the transmission to brake going downhill in manually selected second gear is
no problem, and you can select second at ANY road speed and it will engage. Hope this is food for thought! Brent Covey
------------------------

The torque converter induced crank/thrust bearing failure thing is a theory
I am not comfortable about. I am going to get a definite answer on exactly
whats going on regarding converter fill pressures very soon for the net.
I'll be sure and keep you posted on what I find out, my feeling is its not
the transmissions causing this.

What I do know for sure is the ATRA 'test' frequently quoted is of no value,
and that the 1966-'67 versions of our transaxles with the switch pitch have
a different front pump housing with a bypass built in to prevent pressures
from becoming excessive- which was left out in later versions. It appears
that the second design non- switch pitch version has a modified pressure
regulator that reduced converter fill volume when the pressure boost in Low
and Reverse ranges engages, but I have to establish if this is in fact the
case, and by what amount, and if theres any mitigating factors that could
cause a problem, such as shift kits with high pressure springs, or certain
cooler configurations that could boost pressures to a high degree. I also
cant imagine that in rad oil cooler sitting still for 240 psi.

Even at the very highest boosted pressures, its about 1/3 the amount of
force imparted to the thrust bearing by a clutch pressure plate on a manual
transmission car, and everything I've seen thus far seems to point towards
bad machine work on the thrust journal/bearing as the cause of the problem.

I will find out the definitive answer and let you know. Brent Covey

The Turbo 400s pump pressure is unrestricted, and can exceed 150 psi.

This is what I am hoping to address- the transmission pump output ranges
from below 100 psi to about 230 psi dependent upon gear range selected and
throttle position. The torque converter 'fill' circuit is tapped off this
main pressure line. Oil in this circuit fills the converter, returns from
the converter, is then sent to the oil cooler, and returns from the cooler
to lubricate the transmission internals.

The 1964-1967 THM 400/425's have a bypass built into the front pump to
control pressures in this circuit, it opens when they rise above a certain
quite low pressure, and this was eliminated sometime after 1967, perhaps in
1968 when the switch pitch feature was also discarded. It appears what the
second design unit does is do away with the pressure relief bypass by
limiting the VOLUME of oil tapped off the mainline circuit instead of
controlling the pressures directly by exhausting excess. This means even if
mainline presuures are extraordinarily high, they are restricted in flow and
pressures do not likely become excessive in the converter fill circuit, as
theres no restrictions down the line. It appears the movement of the
pressure regulator valve (on second design units) as it is boosted to higher
pressures covers the port that feeds the converter to restrict flow into the
converter and modulate the pressures somewhat.

This is analogous to your garden hose- the city water pressure is very high,
but if you just turn a trickle on its not high pressure at the end of the
hose. If you bend the hose to stop the flow, the pressure will eventually
rise to match city water pressure as the pressure builds over time behind
the kinked hose, and you'll get a spray. It appears that as the pressure
rises in the transmission, the flow INTO the converter fill/cooler cicuit is
restricted by the motion of the pressure regulator valve to limit the volume
as mainline pressures rise. This effectively keeps the converter from seeing
very high pressures, it apears.

What I'm going to try to measure is what exactly are those pressures in the
converter fill circuit going IN to the converter, under normal
circumstances. Nobody thus far as far as I can determine has ever done this,
which is a lot of why the myth persists. My feeling is they aren't anywhere
close to mainline pressures at all. I have this view especially after seeing
many in rad cooler tank failures where the transmission was not able to put
much fluid in the 15 psi antifreeze, but the 15 psi antifreeze very
effectively filled the transmission. Some of this can be attributed to the
15psi coolant remains pressurized for some time after the
engine/transmission pump stops, but if the coolers had 150 psi etc. in them
the rads would fill with ATF fast if the trans coolers ruptured, which
doesnt seem to happen. I think the converter is seeing quite lowish fill
pressures even when mainline exceeds 200 psi. I do find it hard to imagine
the cheap in-rad cooler tolerating 200+ psi, or the torque converter seal
working at 4000 rpm and 230 psi pressure also.
 

 As I recall, (and could be wrong), the crank is designed to resist up to 225 psi, so 150 psi from the pump converts to about 260 psi pushing on the crank. In other words, at high rpms and full load, the crank bearing is maxed out.

The thrust surface of the crank and main bearing are designed to tolerate a
manual transmission clutch pressure plate spring load. This can be about 750
lbs, but intermittently. For continous duty, I am not sure what it would be,
I've read the same numbers you have Mark, but havent seen anything from GM
regarding this. The crank thrust surfaces can tolerate a tremendous amount
of loading if they're put together right, and put together wrong, they can
fail with NO end thrust applied from that pot lid thing, where they get
stuck together and grind away on thier own like a steaming stove pot lid
gets stuck on the pot when you turn the heat down. Theres no end of ways to
screw up machine work, line boring main bearing bores 'downhill' tips the
crank and forces it to thrust against one end, as can a slightly bent
connecting rod, a tapered journal, or a variety of other mistakes that are
easy for shops to make.

It does appear that if you seriously restricted the cooler circuit flow
(like 90%) in the second design transmission you could conceivably have
pressure build up as the oil backed up between the converter and the
restriction, and this would eventually reach mainline pressures in second
design transmissions. The reduced flow would have other effects as well of
course, you wouldnt have oil cooling, and the transmission lubrication would
slow to a trickle and likely cause problems in the transmission rotating
group. I dont think a series mounted extra air to oil cooler like many
people use would offer a resritiction.

After I have tested a few THM 400's and 425's (I also have a couple switch
pitch units with bypass for comparison) under normal conditions, and with
various changes (shift kit regulator spring, restricted cooler lines flow,
in Low or Reverse for pressure booster actuation, with modulator
disconnected for maximum mainline pressures) I'll have a better idea what
actually is happening. We'll have an actual number.

I do not think, as AERA maintains, that the torque converter fill circuit
operates at a pressure equivalent to mainline pump pressures in short. If it
turns out it does, thats going to be a real eye opener for me. In the event
it does, it begs the question of why they all dont blow up in any event. The
GMC runs the same pressures the cars do afterall. It also seems that crank
problems are a lot more common on rebuilt engines than original factory
installed engines. There have been several engine rebuilders association
bulletins cross my desk in the last ten years related to thrust bearing
failures in other engines that do not have THM 400's and in fact, arent GM
vehicles at all. As far back as the 1970's publications like Trailer Life
were advancing the ballooning (literally) converter theory to explain 454 MH
chassis crank problems.

For now, I'm just building the rigs to test every combination I can put my
hands on, and will publish what I find out so that in the event the
pressures are not the problem, you'll have some data good enough to refute
absolutely any AERA or other information to the contrary. If I confirm the
AERA information, it means we can try to find the explanation for why this
isn't occuring across the entire GM range of cars, every single 1965 to 1980
Cadillac RWD model (and many other Cadillacs) uses a THM 400 and nobody has
ever heard of a thrust bearing failure in one. It may turn out to be some
common parts combination occasionally doesnt stack up right, or shift kits
cause it, or the problem originates in the engines after bad rebuilds. We
just dont have enough good information to prove it one way or the other yet.

Should have something to show you in about a month. The reason for the test
rigs is any test port must not in itself introduce a flow restriction, and I
need to measure the pressure feeding the converter as well, not just the
converter return which is the port to the cooler, so I have to make some
adapters.

Basically, the essence is this whole transmission thrust failure situation
is at best a theory, nobody has any factual information from direct
measurement, and AERA had a big axe to grind on thier unscientific test. I
feel theres a lot of failures that originate from other sources, and perhaps
all, being erroneously blamed on the transmissions. I would like to make the
information availible so that if people subsequently have problems with
engine failures they can put the blame where it belongs, and suspect it may
turn out to belong to the machine shops, or in some cases other factors like
dud harmonic balancers etc. I'll let you all know what I find out soon.Brent Covey
 

Is there any merit in rebuilding our transmissions BEFORE they fail.  If so, when?

There's no reason you usually would want to make a pre-emptive strike on a
transmission. Automatic transmissions are terribly reliable things.

I am of the opinion that many and perhaps most automatic transmission
overhauls done by chain type transmission stores were probably unneccesary,
and most transmissions had smaller problems that could have been corrected
by fixing a particular transmission part or component, often without
removing the transmission from the vehicle. Its very ordinary for the
transmission shop to pull your transmission right out, diassemble it into
little greasy bits on the bench and then show you the disassembled unit and
pressure you into getting it overhauled, regardless of the problem. $1500+
later, its back in the vehicle in about the same condition it came out, with
a $30 repair of the actual part that was causing the problem. This has
always bugged me no end, talk about a license to print money.

Its very hard to predict service life of an automatic transmission. Even a
perfectly assembled top quality unit might suffer a cracked metal part or
pick up a tiny piece of debris is a bad location that could cause problems,
but this is very rare. The general rule is anything that works good today
should continue to work well in future, so the fact that you haven't had a
problem is a good sign. A well assembled transaxle like the original should
last 400,000+ miles quite easily if driven with consideration, and about
half that long if driven without a thought to its welfare.

The life of the clutches and bands is infinite if they are applied at
correct pressures and on time, and the seals are subject to cumulative heat
damage. There is no real normal lifespan for seals, they'll work forever if
they remain soft and pliable. Each time the seals are exposed to very high
temperatures and the longer the duration of these exposures, the faster they
harden, but this doesnt start to be a real factor until above 250F pan temps
which are rarely encountered. This causes damage that acculmulates over
time, getting it smoking hot (over 300F) once is not usually going to cause
much damage, but if this occurs a few times, and the transmission
occasionally spends time at significantly elevated temperatures eventually a
seal may harden or distort enough to permit a clutch failure. The best
defense against this is driving style, the heat is generated in the torque
converter and by minimizing the time the engine spends at speeds below 2500
rpm at heavy throttle (essentially, speeds where the converter is able to
'slip') you can reduce significantly the operating temperatures of the
transmission. If you are climbing a long grade at 3500 rpm and full
throttle, the transmission does not increase temperatures very fast, but
climbing the same hill at 2000 rpm would overheat the transmission in a
matter of minutes. This is one more reason I lean towards the 3.42-3.70
final drives, it keeps the engine turning just fast enough to get beyond the
risky zone in top gear in most driving.

The bearings and bushings have very very long lifespans and generally will
outlive the remainder of the transmission so they are seldom going to cause
any trouble. I have never heard of a chain drive failure, although some are
loosening a little by 100k. I think Caspro may be the only game in town for
a new one, if Cinnibar doesnt offer them.

The hydraulic systems like the valving in the valve body, and the governor
can occasionally stick, pick up dirt or get burrs on them as they wear and
can cause small problems that *if diagnosed in situ* can be corrected
without removing the transmission. Converters seldom fail, and oil pumps are
very reliable, and if they fail, the transmission needs to be removed to
replace them.

If you are pretty certain your transmission has suffered repeated
overheating (venting oil from the vent, and burnt fluid and numerous leaks,
and a vehicle history that suggests extremely hard use) you might consider
pulling it out at your leisure and putting a soft parts kit in to preclude a
failure. This might be 2% of the GMC's on the road. For the other 98%,
theres not much reason to worry about it. The little edges of the valve body
gasket you can see in the oil pan is a good indicator of past overheating,
it will be very brittle and little brittle pieces can be snapped off if you
pick at it with your fingernail on transmissions that have been very hot in
past. If its still flexible and bends probably the reast of the seals are
just fine, the gasket is more sensitive than the internal seals.

A good thing about the THM 425 is it seldom fails in a mode that stops you
dead in your tracks. Even with very serious problems a transaxle with a
substantial failure usually can limp under its own power to a service
facility by driving in a range that does not require operation of the failed
part (hopefully you have options beyond 'reverse'). Most pending failures
also give a few hundred miles warning before they quit altogether.

To give you an idea of what GM does to ensure strength in the THM 425,
theres a test done on Proving Grounds. It consists of revving the engine to
full throttle and shifting repeatedly between "D" and "R" on dry pavement
ONE HUNDRED cycles at ~1-2 second intervals. About the fifth cycle, the
tires catch fire. At the end of the first 100 cycles, the engine is left to
idle to cool the transmission a few minutes and the test is repeated another
100 cycles. This goes on until theres a failure, which can take forever to
happen. I have a movie of a 1959 Chevy with Turboglide doing this test (it
passed) and there was a GM ad on TV about a dozen years back showing a FWD
Olds Cutlass doing it as well. They are really tough!

Generally as you can see, theres no reason to worry or to service a good
working unit until theres a problem. If you have the engine out and
servicing it is convenient, it might be an idea if it has very high mileage,
say over 200k, or evidence of abuse/overheating but most shouldn't require
any attention. I would absolutely consider it vital that any time the
converter is accesible that the seal for it be changed. Watch out for that
AAMCO scenario I had mentioned earlier where they drop it and offer a 'free
estimate'. When its laying in peices on the bench, you're pretty much
commited to an overhaul for big bucks. Transmission failures can generally
be pinpointed with excellent precision if they're still in the vehicle. I
still get about one email a month from the guys with old Cadillacs with
shifting problems with the old 4 speed HydraMatics used from 1940-1964, and
most can be diagnosed accurately by e-mail, so its not hard at all when
you're right there under the vehicle. Hope this is reassuring,

Brent Covey

 [INDEX] | [SUGGESTIONS]