In a past life I built race engines. I also did allot of old restorations. The most problematic one was a 1924 American Lafrance "jug head" Motor. 17 years later it is still running just fine. It took ALLOT of time to repair all of the cracks but it was done.
There are any number of ways to repair cracks in castings. Some make a long lasting if not permanent repair some are just "stop gaps" one I think should show promise would be pressurized ceramic seal in the water jacks. The other would be pull the seats put crack repair pins in then put new valve seats in. These are ordered by size. It does not have to be a "Cat" part. Different seat materials combined with the pins may correct for the problem.
20 year ago racers were still cutting and welding cylinder heads together to get what they wanted.These heads were then modified for better flow. these heads were both cast iron and aluminum. Some of the builders finally got sick of this and started casting their own with all of the port modifications built in. That being said it was found that these new heads could be improved even further.
All of these "trick" heads are very costly. Race engines have some pretty impressive failures. There are people out there that can repair heads that you would think could never be fixed and do it for less than the cost of a new one.
All of this being said a little bit of research should yield someone who can repair them.
As far as the Alternators, I can have anything wound for variable frequency. This would allow 8 throttle speeds.

OK, if we have a control system with defined steps, that seems likely...but the critters I've spent the most time on had a throttle that had no defined steps, just a smooth (hopefully) arc from where the contactor dropped in and out at the bottom end to the highest, and controlled the RPM. There were transition steps within, but essentially the loco speed was controlled by RPM.

One case I will never forget.......a GE with siderods from a dockyard that had a hydraulic system linking throttle to engine. When the hydraulics failed, it was kludged with airplane wire and pulleys. As the wire stretched, you'd haul back until you hard the contactors close, then reach up and pull down on the wire running just below the ceiling to rev the diesel...........

Could be my experience is limited to older critters and more modern centercabs - say, post 1950 - had notched steps? If so, then we may have a workable solution at affordable cost relative to benefit in the works.

Let's not underestimate what it takes to reengineer a new head for something like this and set it into limited production....this isn't just a flatish hunk of metal that seals up the end of the cylinder as a front head on a steamer would be, but one with a myriad of cores and passages and valve seats with very interesting heat and cool cycles to be dealt with. If the rest of the blocks and parts were still available, maybe worth it. With nothing available and everything highly depreciated, unlikely.

If Jack can set up a relationship with a shop that can and will do reliable quality work on really repairing old heads.....great, it would be of a lot of use to us all. There's a lot of variability of quality out there in custom small shop work, and none of it is inexpensive, and probably nobody would guarantee their work.

dave

According to my roster books, only 348 44-tonners were built, with not that many actually going to Class Ones (B&M, PRR, WM, etc.). I would suggest that the 44-tonner gets a disproportionate amount of attention and affection from railfans from a combination of its popularity with modelers and its prolonged existence on "underdog," down-on-their-luck railroads where the best explanation for lack of change is inertia and/or lack of money. Most of the places I saw 44-tonners and 45-tonners survive into my railfanning days certainly fit that description--the Claremont & Concord, various AC&F plants and scrap yards, etc.

I've been retired for more than a year from Tyburn RR, which used Hoboken Manufacturer's RR's old 44 Ton GE 700 as their 400. We got Caterpillar parts from the local dealer, Giles & Ransome in Bensalem, Pennsylvania. But, we could not tell them that the order was for a D17000 in a locomotive, since "Caterpillar doesn't make railroad parts for its dealers to sell". However, the same kinds of Caterpillar D17000 engines were used in packaged electric generator sets, and Caterpillar does make parts for them.

The above may explain why modern Caterpillar powered locomotives aren't very popular, either.

We also had a GE 45 Ton loco which came with Cummings engines. It had 1 bad fuel pump, and McHugh Bros. found that Cummings wanted more to make a replacement than General Motors wanted for 2 complete new Diesel engines of their make! But, the main generators have only one bearing, the other end is supported by the Cummings engine, and the bolt pattern matches only the special proprietary shape of a Cummings engine. Can you imagine the cost and complexity of designing and making adapters to mate those generators with some other make of Diesel engine? General Motors can, and they sell the proper adapters. If they're not in stock at the time, they'll give you a copy of the plan so you can have it made elsewhere, which is what McHugh wound up doing!

You just missed out on getting parts from a 44Ton GE that Southern Rwy. of New Jersey recently scrapped at Winslow Jct.

One advantage of the 44 Ton loco versus the 25, 30, 45, and 50 ton and some other similar sized critters is that it is allowed 35 miles per hour, the others are limited to 20 m.p.h. before the traction motor armatures fly apart.

Having spent a great deal of time this summer getting intimate with the GE 44... let me comment.

The mere fact of being frozen up says nothing about the health of the components. Two options: #1 tear the engine down lovingly and properly, removing grime, scale, rust, pitting etc. so it can run at peak performance for a long long time. #2 use voodoo juice to do the bare minimum to free up the engine, then run it til it breaks, then get another locomotive. Seriously I would not be in any hurry at all to "get the engine to turn over". In fact I would want it to NOT turn over. The last thing I'd want to do is drag rust, scale or other garbage across cylinder liners and bearing surfaces, which are otherwise (at this point) uninjured but merely dirty. There are few parts of a D17000 (aside from crank and block) which weigh more than 200 pounds. You should be able to carefully disassemble it and clean it.

P.S. If you need to remove the engine hood, IT UNBOLTS. Put the torches away.



The D17000 mounts on a skid with the GT-555 generator. The D17000 is cast specifically to mate to a generator. The GT-555 is made only to mate to the D-17000 and relies on its rear main bearing (single bearing design). It would be a monumental engineering task to mate it to another prime mover or gearbox. You would be better off swapping the whole skid with an engine/genny combo made to work with each other. Then you would only have to contend with throttle (really RPM select) cable, exciter and belt drive connections - and if you were smart you'd position them the same as on the factory skid, and with the correct pulley sizes to net out the right RPM. That way it's a drop in replacement instead of a kludge.

It wouldn't really matter if it was AC or DC but the GE 44's operating traction voltage is 250VDC, so you'd want a 220VAC alternator rather than 480VAC. You'd want it to be a good match to the GMG-140 exciter, or bypass the need for it (careful here).

I have heard lots of stories about this sort of thing being screwed up spectacularly, leaving a failed "science project". You have to have VERY good design discipline, and deep respect for what GE originally built. And you will have Z-E-R-O support from ANYONE, so you better be darn good.


I'm always concerned when I hear about people talking about cast iron welding... this is a VERY deep subject, and it's difficult to know how deeply someone has gone. For just a glimpse into the complexity of high-end cast iron welding, http://www.locknstitch.com/CastIronWelding.htm

They may be able to help with the underlying problem, because it may be due to stresses they can work out of the piece via control cooling. We know that these things didn't fail this way when new. Cat would have fixed the design if they did.


Well OK fine, but that's just the kind of "scope creep" that turns these science projects into disasters.

Under the cab you have a GMG-140, an auxiliary generator and exciter on a common shaft. That is belt driven. You also have a Gardner-Denver ADS compressor.

The compressor is, obviously, easy; however you need to rely on the existing governor or devise a new way to sync the compressors so they run in unison. Otherwise one compressor will do all the work.

Since you're providing the main generator you get to decide what it needs for excitation.

The auxiliary generators are trickier than you think. There are two areas of significant design concern. #1, BOTH aux gens must be disabled whenever EITHER engine is being cranked. Otherwise one engine will try to "help crank over" the other electrically, and you don't want that unless your aux gen is extremely tough. The easiest way to accomplish this is to leave in the 60B/E circuit, and the A1/A2/A21 contactors. #2, something may need to be done to balance both aux gens, so they are not "fighting" each other. This isn't magic. GE does this with circuit 29, a "bias" coil running between both regulators.


Not a factor. Everything EDM wants to do can be contained within the skid. He wants to put the aux gen and compressor's functionality onboard the skid, well fine, leave the originals in place. They're not in the way. Make 1 air connection for the compressor and 1 for the governor... then 2 connections below the A1 contactor for the aux gen, 3 for the gen, 3 for the exciter (if you even do it that way)... and yer done. Then if you want to roll back to a real D17000, that could happen in about 8 hours. Again, it's about understanding and respecting the original design and building something that plays well with it.

On the skid, anything goes. You could use a stack of Chevy Volt battery packs recharged by a Wankel engine, for all I care. As long as its basic connections are the same as the D17000 skid.


No. Not at all. You simply rev the genset up and down as you wish. You would probably be changing the engine governor so instead of holding [1800] RPM, it would hold [your desired] RPM based on throttle position. The rest is exciter design. Rectify to DC right before it leaves the skid to go into the traction motors.


Yes, that's how most GE 44's work. There is a throttle shaft that pushes up and down on the engine governor, which selects RPM based on "throttle" position. GE 44's were available in MU versions, however I have only ever seen three of them with M.U. They are all at Rio Vista Junction. I don't know exactly how it works.

Your skid contraption should be designed to accept a sliding rod as its RPM/power selector, because that's probably what it's going to get.

Of course if you want to scrap out a GE 44 and use the hulk to build a "blank slate" new design of locomotive... good luck with that LOL. Most outfits have enough trouble with an engine swap.

Inching closer.......

Just for a bit more information, I've checked out a few sources of older diesel engines and discovered a lot of marine engines were paired with gear reduction units in the 150-350 HP range........so, presuming you do build a end cap and bearing for the generator shaft that hangs out when the motor is removed, seems like a possibility to use some 30 year old marine diesel plant rather than a 60 yar old heavy industrial diesel and get something that will work....without the need to modify the rest of the ecosystem too much. And yes, I'd cerainly want to mount itn on the skid if at all possible. Life is complicated enough without trying to work up on the frame.

I haven't had the chance to explore the world of available DC generators that might be mated to a modern diesel yet...... maybe there's another direction along that path that could be worth exploring.

dave

One of the mods should make MacDowell's post one of the stickies -- it's that valuable.

Having no luck with the link to the DC genset spec pages. Anybody else have better luck?

dave

Have informed Cummins via e-mail and Web form that the DC genset page is down with a Virtual Directory (likely Microsoft software configuration) error.

Also have asked for whatever technical detail they can provide on DC genset retrofit applications for locomotives, and whether the PowerSuite DVD (available from your local Cummins dealer, but not by mail or download) contains DC genset technical information.

I won't get down there today... but have to contact Cummins on 6BT technical matters in the next several days, so if nothing else happens on this thread, keep watching...

Didn't the fellow who initiated this thread indicate the group in question was on a budget ????

One has to wonder about the efficacy of 'investing' all this money in a locomotive with superannuated controls, wire and cable, radiator, etc.

DPK

EVERY group is on a budget, except for those that don't actually budget, who are still on a budget but they don't know it yet. Some budgets are bigger than others......and the most expensive train is the one you can't run when your only working locomotive is down.

We work from a bundle of reasons when making these kinds of decisions, and what we try to do if we're smart is prioritize all the sticks in the bundle and try to find the most rational and sustainable solution we can afford.

Sometimes there isn't one we can afford. Process starts again with a different set of criteria and an adjusted goal.

That's why I'm trying to get a brain trust going here that will perhaps eventually land on a solution that might not be exactly a "drop in" conversion, but something that's technically doable even if costly, for those who want to keep them running for other than financial reasons entirely. If we're very lucky, perhaps a few solutions that each have different compromises from which to choose, based on affordbility among other things.

Dug out my old manual for a 1946 Alco / GE 44 tonner and copied out the curve diagrams. I'll try to scan and post soon.

dave

Reverse current diodes can keep the two aux. generators from trying to spin each other; I haven't seen a reverse current relay in years. The two compressors can be synced with an unloader, NS-16 comes to mind, just pipe the unloader air signal to each of the compressors.

Agreed, the suggestions I made were 'perfect world, big budget'.

EDM

A lot of good info so far, but with the compressor issue, it looks like you guys are over thinking it. As designed the 2 compressors on a GE center cab work independently, as you don't need to run both engines and their drives are not connected.

The compressors have unloaders built in, when the pressure gets high enough the unloader sends air to a diaphragm in the head which holds open the intake check valves and the compressor just cycles air in and out of the intake. Since the cylinder can't build any pressure the exhaust check valve stays closed maintaining the system pressure. When the pressure drops, the diaphragm deflates and the intake check valves operates normally. This is the way most engine driven reciprocating compressors are set-up.

As far as D17000 parts go, as long as you don't have a first generation engine (crab bolt held injectors and exposed layshaft water pump) you can get parts from Cat, just tell them it is off a D8 dozer, a gen set or you are fixing a D13000 off a D6, I forget what else came with a D17000 or D13000 installed. The D13000 is a straight 6 engine that uses most of the same parts as the D17000 (exceptions are crank, cam, block, fuel rack and the fuel distribution pump IIRC). Also this means keep an eye open at the equipment auctions, as you may find a used engine for parts there. IIRC Claremont & Concord has made new water pump layshafts for the earlier engines and were selling them a few years ago, but that was before they sold their 44 tonners, but if they are out of stock, they may still have drawings. 3 or 4 years ago we were checking on the part prices in case we wanted to rebuild our spare block and were able to get pricing (not cheap, but this is Cat we are talking about) for most of the wear parts, IIRC the threaded body injectors were one of the hard to find items. Unfortunately our batteries froze and delayed that operation.

Oh, so you can learn from the previous owners of our 44 tonner, remember to not reuse the copper gaskets and water seal rings under the heads, once used no matter how much RTV you use they won't re-seat correctly. It was found once we replaced these the water/oil cross contamination problems were solved (contamination went both ways).

Rich C.
Ct Eastern RR Museum

Rich, that's an interesting comparison of the cat D17000 and D13000 series engines. IIRC, the Tourist line at Abilene KS has operating examples of both. They have a 25-ton locomotive crane with a D13000, and a very early model GE 44 ton which I believe still has both original D17000 operable. One of the officers there is retired as owner of a Diesel repair shop, and is still active in maintaing their locomotives.
Dan Rohrback

Go with an AC main generator if you can get one. Three phase AC is effortlessly turned into workable DC by six diodes. That leaves you two problems:
#1 you'll need a way to crank the engine over, since the DC generator could be used as a motor, but now the AC alternator cannot.
#2 (but this is a problem you will have with ANY replacement generator of any kind): having it electrically match the excitation put out by the GMG-140 exciter... or figure out a new and different way to excite.


If your requirements demand a state of the art locomotive, you are in the wrong thread if not the wrong site. As far as "all this money", maintain an EMD anything and get back with me.

Dealing with a 1943 GE 44, there's nothing really wrong with the wire, and the controls are serviceable, there's nothing "unobtanium" in any of the electrical or running gear. At the end of the day, what you have is a 44-ton locomotive, which is an absolutely excellent size and design for the passenger-hauling needs of a modern railway museum.

Now that SMRS's GE 44 is tuned up, we can run every car on the railroad, on any part of the railroad, on a single engine. With two, it's effortless. So we have self-provided "backup", which gives our Board of Directors confidence to schedule and market revenue service, for more revenue. I was around SMRS in the 80s. Years of running a Plymouth, we longed for a GE 44. A founder said "it's like a cockroach, cut it in half, it still runs." We fantasized it as the perfect solution to our troubles. We were correct then and are still correct now.

The only real downside IMO to the GE 44 is the D17000 engines. Solve that problem and you're quite good.


All this has been done for you by GE, it's already there in the design.

A reverse current relay is not QUITE the same as a diode. There are some subtle differences. An RC relay will trickle back some reverse current, which your regulator might find useful to start up. Also, your RC relay can be interlocked with other stuff - a diode cannot. On ours, the previous owner's diode upgrade neglected the circuit which isolates the auxiliary generators during engine cranking, causing numerous aux gen and regulator meltdowns before it was caught. This sort of bug is TYPICAL of any sort of cowboy redesign, and the more you want to change, the more problems like this you WILL run into. The only question is whether you have the stamina to fix them all, or whether the project will get abandoned as most have.


Almost. They do not independently make the decision of WHEN to cycle on and off. That is made by a SINGLE compressor governor for the entire locomotive, which commands both unloaders to unload (or load) at once. That way both compressors pump (or don't) at once, and you don't have one compressor never pumping because the other one cycles a fraction of a pound sooner. Ask the trolley guys why multiple compressors are synced.

A lot of things we take for granted seem like magic. In fact, they are the product of careful design by the builder.