Hydraulic Cooling Fan Issues

[Truthseeker4449]

If memory serves this bus has 9 pin. I know my other bus definitely had 9 pin, they even left the telemetrics harness plugged in and lopped off the tablet connection above the drivers area. Both buses do have ABS, tho apparently the MVP is not configured correctly. According to the dealer, the control module thinks the bus has 3 axles, they're supposed to take care of that too. Maybe they'll reprogram the tire size while they are in there.

MD3060s are definitely still being built, the last fleet I worked on have quite few and my current job does too. Definitely don't want to mistake one for a 4000 during a service, 26 quarts of expensive fluid will completely fill a MD3060 and spill out the breather.

[Booyah45828]

Originally Posted by cadillackid

if you look at how WTEC-III works it talks the J1939 language... not saying allison didnt run some bastardized version.. however J1708 and J1939 are physically different.. the levels are different.. you cant intermingle J1939 and J1708 on the same pair of wires.. and WTEC-III TCMs only had CAN-1 (not the second set of pins for CAN-2 that the 4th gen has).. the WTEC-III shift panels come up as J-1939 controller address 5.. the ECM is 1.. and ive witnessed this myself by tapping into the trans links..



on my IH the connector that goes from engine to trans is Pure J1939.. ive decoded it, ive measured it.. and that is a year 2000... any allison that got connected to the link in said year was WTEC-III.. in fact IH released the J1939 Trans connector and software for the NavPak in 98 (ironically coinciding with the year WTEC-III was released)...



WTEC-II was J1708 in its root with a bunch of proprietary PID's used to handle better transmission diagnostics and the language between the TCM and shift panel... Most WTEC-II integrations used a mechanical throttle but some like IH put the trans on the J1708 interface on the 466E.... and it worked sort of OK for the tech at the time..



after all when the world-series came out in 94 the whole realm of electronic shifting automatics was a pretty new thing..

Is yours a 1/2k or 3/4k? 1/2k was 1939 native, and I don't think they were ever 1708. 3060 was 1708 native originally, and it wasn't until after CAN networks become mainstream in early 2000's that the 3/4(3060) went to j1939 and CAN.

And to say allison has a bastardized 1939 is funny, you should see how cummins did it in the mid 2000's. Certain interfaces can't even recognize it because of the odd baud rate, but it's there, and is "1939".

[Booyah45828]

I don't think they're building 3060's anymore. They're all 1/2/3/4k's with the correct suffix, like pts, hs, mh, etc.

They did that in the late 2000's.

[Truthseeker4449]

Originally Posted by Booyah45828

I don't think they're building 3060's anymore. They're all 1/2/3/4k's with the correct suffix, like pts, hs, mh, etc.

They did that in the late 2000's.

Maybe that's what it is, I don't think they actually changed that much tho. Different strength components and guts for different applications sure, but it wouldn't surprise me if many of the parts would still interchange. I've taken a few control modules out (the integrated pan and valve body assembly) at work, but admittedly not something 20 years old yet.

[cadillackid]

Originally Posted by Booyah45828

Is yours a 1/2k or 3/4k? 1/2k was 1939 native, and I don't think they were ever 1708. 3060 was 1708 native originally, and it wasn't until after CAN networks become mainstream in early 2000's that the 3/4(3060) went to j1939 and CAN.

And to say allison has a bastardized 1939 is funny, you should see how cummins did it in the mid 2000's. Certain interfaces can't even recognize it because of the odd baud rate, but it's there, and is "1939".

my point wasnt about my specific transmission.. my point is about the transmission INTERFFACE on my international that is DESIGNED to hook up to allison 3000.. (3060 was common in school bus).. and its J1939 and it came out to coincide with WTEC-III...



my allison 1k wasnt even out when the interface was released.. only the 3/4 were electronic in that era.. you are correct the 1k/2k are only WTEC-III.. (till the 4th gen).. and yeah a lot of the early stuff runs on 250k, (my IH is 250k even talking to my 4th gen allison)...i dont think the trans interface went 500k 1939 until ECM-II (think EGR era)..


yes the J1939 was optional on the 3126 as you've said.. in 2001 you could still order a 500/600 or a manual and with no ABS,(or standalone ABS) thered probably be no reason to install it..



but having a WTEC-III allison would be a good reason for 1939 to be present.



as far as bastardized protocols?? yeah theres a lot of that.. WTEC-II was such an animal... and IH of course did it.. (ever noticed how many PID's that IH pushes through that arent in the J1587 spec?)..


but I havent seen allison do that on WTEC-III.. every one (WTEC-III allison)that ive sniffed the link on.. (back to an 00 A3RE with a cummins ISC) all ran straight 1939.. some 1939 interfaces provide more data than others.. PGNs for EEC 1 and EEC 2 are the most basic which I saw on the cummins.. IH sends a bunch of other stuff ... the allisons send a good amount of data (they only respond to the inouts that are turned on for their TCM package)...

[Booyah45828]

Originally Posted by Truthseeker4449

Maybe that's what it is, I don't think they actually changed that much tho. Different strength components and guts for different applications sure, but it wouldn't surprise me if many of the parts would still interchange. I've taken a few control modules out (the integrated pan and valve body assembly) at work, but admittedly not something 20 years old yet.

Yeah, I don't think they changed much. Cases and such all looked the same.

You'd have to talk to a rebuilder for all the details. The rare few that we have internal issues with are swapped with remans from Weller.

MD3060, HD4060, B3/4/500, etc were designed between the mechanicals and world series transmissions. As such, they have they're own method to naming, which doesn't follow the transmissions designed before or after.

[Booyah45828]

Originally Posted by cadillackid

my point wasnt about my specific transmission.. my point is about the transmission INTERFFACE on my international that is DESIGNED to hook up to allison 3000.. (3060 was common in school bus).. and its J1939 and it came out to coincide with WTEC-III...



my allison 1k wasnt even out when the interface was released.. only the 3/4 were electronic in that era.. you are correct the 1k/2k are only WTEC-III.. (till the 4th gen).. and yeah a lot of the early stuff runs on 250k, (my IH is 250k even talking to my 4th gen allison)...i dont think the trans interface went 500k 1939 until ECM-II (think EGR era)..


yes the J1939 was optional on the 3126 as you've said.. in 2001 you could still order a 500/600 or a manual and with no ABS,(or standalone ABS) thered probably be no reason to install it..



but having a WTEC-III allison would be a good reason for 1939 to be present.



as far as bastardized protocols?? yeah theres a lot of that.. WTEC-II was such an animal... and IH of course did it.. (ever noticed how many PID's that IH pushes through that arent in the J1587 spec?)..


but I havent seen allison do that on WTEC-III.. every one (WTEC-III allison)that ive sniffed the link on.. (back to an 00 A3RE with a cummins ISC) all ran straight 1939.. some 1939 interfaces provide more data than others.. PGNs for EEC 1 and EEC 2 are the most basic which I saw on the cummins.. IH sends a bunch of other stuff ... the allisons send a good amount of data (they only respond to the inouts that are turned on for their TCM package)...

I know I've used 1708 on wtec3 ecu's w/ md3060s and iirc a dt466e(coulda been a 408, but I don't think so). It was 6 or so years ago on a 2001 IH straight truck used by a fertilizer dealer, and it only had the 6 pin deustch connector under dash. Original complaint was the truck would run rough and shift display would go dark after a couple minutes of being on. Connecting the scan tools revealed 113 in the ecm and no comms to the allison.

Being a fertilizer truck, the battery box was full of corrosion and the separate ecm fuse wire was corroded nearly off. Once fixed, I still couldn't communicate with allison doc. In the end, I had to deselect smartconnect, and then select usblink2 and j1708 for it to work.

As I said, it was definitely a 3060, and with the 6 pin connector it shouldn't have had 1939 present at the dlc.

[cadillackid]

on IH they have passthrough.. its handled by the ECM.. you connect up DOC to whichever connector you have and the IH ECM handles it..



ever noticed how even when you connect servicemaxx to 1708 that it shows J1939 devices?
my 99.5 (2000 MY) has a 6 pin connector under the dash.. yet J1939 connector was present (and in use) when i swapped in my allison i found that link and connected it.. works great..


when I hook up my USBlink2 to the J1708 (6 pin) connector under the dash it shows my transmission online as a J1939 device and has the TCM module ID.. when I hook up an electronic shift pad (yes you can run electronic shift pads on a 1k/2k contrary to popular belief(they wont shift the hard gears but do shift the ranges)..



the shift pad shows up on the J1939 data bus.. even though im plugged into the J1708 6 pin connector under the dash....

[Booyah45828]

So it will use 1939 between ecm and tcm and shift pad? Never knew that. I just assumed that without can bus present at the dlc it didn't have a can bus.

I didn't know you could add a shift pad to a 1k. I never thought of doing that. It makes sense because IIRC allison offered a t-handle with tapshift features, but I've never seen one in person and don't remember if it could be used with a 1/2k trans, or was like the 3/4k and the t-handle didn't actually shift anything. I think later duramaxs had a tapshift option on the shifter too, but like you said, you still need something to put it in the p/r/n/d range, that part is always mechanical on a 1/2k afaik.

[cadillackid]

Originally Posted by Booyah45828

So it will use 1939 between ecm and tcm and shift pad? Never knew that. I just assumed that without can bus present at the dlc it didn't have a can bus.

I didn't know you could add a shift pad to a 1k. I never thought of doing that. It makes sense because IIRC allison offered a t-handle with tapshift features, but I've never seen one in person and don't remember if it could be used with a 1/2k trans, or was like the 3/4k and the t-handle didn't actually shift anything. I think later duramaxs had a tapshift option on the shifter too, but like you said, you still need something to put it in the p/r/n/d range, that part is always mechanical on a 1/2k afaik.

the Tap-shift feature simply sends a range down or range up command in the 1939 stream.. durammaxes are different animals.. they work on GMLAN which was GM and allison's doing when the original 1000s came out for duramaxx.. there is a way to do it with a physical "wire" also.. if you look at an allison design manual.. it refers to most all inputs as "wires" buit will find that most of them also have a J1939 Command COunterpart that will perform the same function.. (as long as that input is enabled in the TCM).


on a 1k/2k the T handle is "god" .. not only does the TCM read it but obviously it is making physical valve changes in the valve body that no computer can override. if i put a 1k/2k in '4' then even with the TAP feature there is no way i can get 5 or 6.. if i send the J1939 PGN for it then i still get a '4' back for the requested max gear from the TCM. I think in the duramaxxes the DIC display showed something like "shift not allowed" as GMLAN has its own custom data stream..



from what I have seen on the 4th gen stuff the GMLAN duramaxx stuff is that those TCMs are even numbered, like A40, A50, T50.. whereas the medium duty TCM model numbers were oddslike A41, A51, A53.. I suspect the differences were in the CAN 1 interface.. whether it was phyiscally J1939 or physically GMLAN..


remember J1939 is a form of CAN bus... allison labels the ports CAN 1 (high / low) and on a 4th gen TCM you have CAN-2 (high / low)..


on my 4th gen TCM, CAN-1 is J1939 for talking to my T444E.. and CAN-2 is actually set up as J-1850 (yep good ole OBD2!).. and can be read with any OBD2 scan tool like a "launch".. that "OBD2" port is how i tune it with my EFIlive(essentially a duramaxx tuner running in trans tuner-only mode)...


there is a **LOT** you can control with J1939 into your allison.. I have found they respond differently in cruise mode vs non cruise mode.. you can force them into compltely different shift schedules with a J1939 command..



I have schedule A-B-C-D (what EFI names them).. so I can have a mountain down (say D).. which raises the shiftpoints way up for descending mountains (keeps lockup on as long as possible)...

as an example

[Truthseeker4449]

I somewhat conveniently had a DEF heater hose fail first thing in the morning at work and so had a reasonable excuse to go to the dealer on the clock and so finally got to talk to the technician and get some wiring diagrams.

The original fan controller is not connected to the data network and solely receives information from its own coolant and charge air temperature sensors. The controller gives a power to a separate solenoid block (located separately below the fan motor) to slow down the fan. It has nominally has one speed. Unplugging the solenoid lets the fan run at full speed.

The $4000 update kit is solely a fan controller and harness. The new controller is setup to take data from a J1939 source. Using that data it can create two speeds by pulse width modulating the original solenoid block. The harness is large enough to plug into the existing data line, which it does on my bus. However because the bus does not actually use J1939, the controller is completely useless. The tech confirmed that around the turn of the century is a weird era for communication protocols and says he verified the controller was in fact receiving data.

The power steering cooler is apparently not run with the fan controller, so something else runs that electric fan.

If anyone is curious, a radiator for one of these buses is about $5,000. So make sure you clean them regularly to keep the cores from rotting out. Transit bus radiators will be around $10,000, or at least the one we're trying to get for an El-Dorado National at work is.

Now for the pictures, first is a shot of the location of the controller. The controller is not on the bus, but would be on the side of the valve block under the fan motor. The tech's fingers are right next to the spot.


Now for the bad controller, the left connector is goes to the solenoid.


Wiring diagram for the original controller


Wiring diagram for the new controller

[Truthseeker4449]

So a buddy of mine has been looking into this further and his researching is suggesting these solenoid valves are not Pulse Width. They may actually be Solenoid Proportional Valves that are constant voltage, but variable current. He can't be sure though until I can take the old fan controller apart.

[TJones]

Can you pull manufacture part # off of the component that failed and then look for a replacement? They were likely used in other industrial applications.

Ted

[Truthseeker4449]

Originally Posted by TJones

Can you pull manufacture part # off of the component that failed and then look for a replacement? They were likely used in other industrial applications.

Ted

I didn't look at the module too hard once I saw the sticker was obliterated. Apparently they had a high failure rate, and have been out of production for a very long time. Even the update controllers are out of production. The Thomas part number for the original controller is 61280148.

I've noticed that Sauer Danfoss makes solenoid valves with what appears to be the same connector as what is on the bus, some of which have a controller attached.

This one in particular seems to be a reasonable match and a bargain compared to the useless update kit. I'll be asking for a part number on the solenoid valve to try and compare with, though it'll probably be a Thomas number.
https://www.ebay.com/itm/283753531497

[cadillackid]

it seems the best option is something that can run standalone with a simple thermo-switch so you dont need to involve the ECM.. some of this stuff gets made overly complex when it doesnt need to be.. im all for comouter control but when the fan is just going to be 2 speed anyway theres no reason the ECM has to be involved.. if the computer gave you vairable speed it would make more sense to try and hook up a data link.. otherwise it just complicates things

[Iceni John]

Why is everyone here focused on the electronics, and not the actual fan motor itself? It's all well and good to come up with a work-around for the control circuit, but if the motor itself is bad, then what's the point? I would test the motor first, then also test the hydraulic/PS pump, the priority control valve and the solenoid directional control valve; only when I'm sure they all work correctly would I focus on the electronics. It's like taking your car in for service because it's not running: the first thing to check is whether it has fuel, air and spark, not what codes are showing.

Why does a fan motor have to be controlled electronically? It only needs to run at half-speed or full-speed, and this can be achieved by simple temperature switches, not an arcane Rube Goldbergian digital complication! KISS.

John

[Booyah45828]

Originally Posted by Iceni John

Why is everyone here focused on the electronics, and not the actual fan motor itself? It's all well and good to come up with a work-around for the control circuit, but if the motor itself is bad, then what's the point? I would test the motor first, then also test the hydraulic/PS pump, the priority control valve and the solenoid directional control valve; only when I'm sure they all work correctly would I focus on the electronics. It's like taking your car in for service because it's not running: the first thing to check is whether it has fuel, air and spark, not what codes are showing.

Why does a fan motor have to be controlled electronically? It only needs to run at half-speed or full-speed, and this can be achieved by simple temperature switches, not an arcane Rube Goldbergian digital complication! KISS.

John

Because the dealer diagnosed it as a failed control unit, the replacement unit is NLA, and the updated unit(that's also discontinued) doesn't work with his bus due to the electronics being incompatible.

I guess I didn't feel the need to second guess the issue as being a failed controller......

The reason to have it all computer controlled is modern computer systems already have the multiple temp sensors available, and they also have a neat self diagnostic feature if one of the sensors fail. On an older system, the way you determine if the temp switch has failed is often after the cooling system, transmission, or power steering system has puked it's guts due to overheat.

Originally Posted by Iceni John

It's like taking your car in for service because it's not running: the first thing to check is whether it has fuel, air and spark, not what codes are showing.

That's not the case with today's vehicles. 99% of the time a no start will show codes. I can have the codes read and troubleshot before I could pull the first spark plug now-a-days. Most of your diagnostics is electroinc, where as the old days you'd use fuel pressure gauges, timing lights, voltmeters, etc.

That's why today's technicians struggle with older vehicles. The mindset to repair the old ones is completely different then new ones.

[Iceni John]

Originally Posted by Truthseeker4449

If anyone is curious, a radiator for one of these buses is about $5,000. So make sure you clean them regularly to keep the cores from rotting out. Transit bus radiators will be around $10,000, or at least the one we're trying to get for an El-Dorado National at work is.

Why so much? A few years ago I had Atlas Radiator in Santa Fe Springs CA build a new core for my bus, and it cost only about $1600-ish. It's a custom-made high-performance 6-row copper and brass with dimpled tubes and corrugated fins. If you buy a radiator from a reseller it will cost twice (or more) what a radiator builder will charge you: the little radiator shop down the road from here wanted well over $3000 for one like mine, and that was without the labor to reinstall the tanks with 1/8"-thick gaskets and all stainless bolts and Nyloc nuts, and to then pressure-test it for 24 hours.

John

[Truthseeker4449]

Originally Posted by Iceni John

Why is everyone here focused on the electronics, and not the actual fan motor itself? It's all well and good to come up with a work-around for the control circuit, but if the motor itself is bad, then what's the point? I would test the motor first, then also test the hydraulic/PS pump, the priority control valve and the solenoid directional control valve; only when I'm sure they all work correctly would I focus on the electronics. It's like taking your car in for service because it's not running: the first thing to check is whether it has fuel, air and spark, not what codes are showing.

Why does a fan motor have to be controlled electronically? It only needs to run at half-speed or full-speed, and this can be achieved by simple temperature switches, not an arcane Rube Goldbergian digital complication! KISS.

John

Well the motor isn't bad, that was a miscommunication between the service advisor who didn't really understand the system, but knew that I needed update kit X based on what his technician said. In fact without the controller the motor spins at full force.

Now this is not desirable as this is extra load on the engine (diesel is still well over $4.50 in my area) and it's extra wear and tear on a lot of expensive things. It also sucks up much more road debris that not only chokes the radiator, but can act as a salty sponge that destroys the $5,000 radiator.

Why does it have to be so fancy, no clue. It appears the valve block that modulates the fan speed has fancy solenoid valve to control. That valve may not actually be PWM in the normal sense either and just happens to have favorable response to adding 12v and removing it.

So I can setup something primitive, or something more advanced that mimics the original functionality (which may actually be more advanced than I was originally told), or bust open my savings account later down the road. Or cut my losses and get another bus, but I don't want to do that option.

[Truthseeker4449]

Originally Posted by Iceni John

Why so much? A few years ago I had Atlas Radiator in Santa Fe Springs CA build a new core for my bus, and it cost only about $1600-ish. It's a custom-made high-performance 6-row copper and brass with dimpled tubes and corrugated fins. If you buy a radiator from a reseller it will cost twice (or more) what a radiator builder will charge you: the little radiator shop down the road from here wanted well over $3000 for one like mine, and that was without the labor to reinstall the tanks with 1/8"-thick gaskets and all stainless bolts and Nyloc nuts, and to then pressure-test it for 24 hours.

John

Sorry missed this one. I don't know, I'm not involved in the supply chain or production of such parts, and fortunately I'm not footing the bill for one yet either. Part of it can probably be chalked up these are typically sold to governments that usually committed for keeping the buses longer than a typical car buyer and do maintenance in house. Transit buses in particular can be expensive, I've heard $750,000 for a CNG bus with zero options, not even a fuel gauge. So it stands to reason parts will be expensive and in my experience, many transit bus parts can only be acquired through the bus manufacturer.

Make your money on the parts and service, not the bus itself.

I think there is a place locally that can build radiators, but I've never dealt with them before on that front. Usually the companies I work for just pay up for new ones.