How is the PRR GG1 articulated?

[aegrotatio]

I see it mentioned in PRR GG1 reference material that the frame is articulated which allows it to negotiate tighter curves than it normally could.  However, what does this mean, is there a pivot in the center?  How does the carbody pivot?

 

[oltmannd]

The two pieces of the frame pivot in the center.  The carbody is rigid.  It sits on the frame with "feet" that ride in slots so the frame can rotate and move under the carbody.   There's a good picture of this out there somewhere....  I'll try to find it.

[daveklepper]

This was also true of the New Haven's 2-6-2+2-6-2 EP2, 4-6-6-4 EP-3, EP-4, and EF-3 locomotives, the EP-3 tested by the PRR as the basis for the GG-1.

[CSSHEGEWISCH]

Most pre-WW2 electrics were articulated, NYC's S-motors, PRR's B1, P5, L6 and O1 are the only ones that come to mind that weren't articulated.  An articulated diesel that comes to mind is the Baldwin Centipede.

[samfp1943]

The following link is to a website that has posted some rough mechanical and dimensional drawings of the GG-1 family of PRR Electrics.

http://prr.railfan.net/diagrams/PRRdiagrams.html?diag=GG1_4800-E404453.gif&sel=ele&sz=sm&fr=

[ndbprr]

The engine as stated rides on two sets of castings that each have six drive wheels and a four wheel lead truck.  Not asked but part of the story is the 4-6-0 steam classification on the PRR was the G class.  Hence a 4-6-0+0-6-4 electric was obviously a GG class and the GG1 was the first and last in it.   There is a difference in the castings in that one has a double keeper and one has a single keeper that fits in the slot of the double one.  Then a large pin possibly as big as 12" in diameter is forced in to keep the trucks together.  There isn't a lot of difference between the real thing and the models except that the models eliminated the keepers and pins and made the pivot point at about the middle drive wheel to allow a smaller radius turn.  Another  aside is that the Baldwin Centipedes were failing when new rounding Horseshoe curve.  Designed about the same as the GG1 it turned out they were so long that the blower in the carbody and the air duct to cool the traction motors would get misalligned significantly shutting off cooling air to the motors and causing overheating and shut down of the engines.  Don't really know how they fixed it but they did.  G's never had that problem but if you ever see any pictures of Don Wood taken at the Wilmington shops you will see spring loaded plates on the trucks of torn down G's that pushed against the bottom of the carbody to blow air on the traction motors.  They appeared in Trains many years ago and in pennsy Power by Stauffer.

[Leo_Ames]

Surprised this hasn't been mentioned.

http://www.railpictures.net/viewphoto.php?id=100067&showcomments=true

[Paul of Covington]

     There was a discussion of the GG1 in one of these forums about a year ago that included pictures showing details of the articulation, but I haven't been able to find it.   Maybe someone who is better at it can find it.    There was also mention of the "G1" in Leo's picture.

[Paul of Covington]

   I tried again, and I found it.   Picture of the articulation:

http://ctr.trains.com/~/media/images/online%20extras/gg1%20in%20layers/gg1-12-1024.ashx

  Link to the forum discussion:

http://cs.trains.com/TRCCS/forums/t/182574.aspx

   (I hope it works.)

[MJChittick]

Paul of Covington

     There was a discussion of the GG1 in one of these forums about a year ago that included pictures showing details of the articulation, but I haven't been able to find it.   Maybe someone who is better at it can find it.    There was also mention of the "G1" in Leo's picture.

You're right and I was finally able to find the reply and link I referenced a year ago!!!  Here it is:

I found a link on the Classic Trains website you should find of interest as it contains cutaway drawings of the interior layout. 

http://ctr.trains.com/Online%20Extras/Equipment%20Rosters/2009/04/GG1%20in%20layers.aspx

The last two or three slides show the articulation in great detail.

[aegrotatio]

The G1 (half GG1) is one of my favorite shop hacks.  It didn't really answer my question about articulation, though.

Thanks for all the great information, folks!

 

[Leo_Ames]

Yeah, but it was made possible due to the GG1 being basically two separate locomotives articulated together.

[PRRrailfan]

But where did this arrangement originate?  Earlier Pennsy had borrowed EP-3 0354 from the New Haven for mainline testing, and liked its performance.  And so one of the prototypes, the GG1,  copied its arrangement—the 2-C+C-2 articulated design, but replaced its boxcab shape with the streamlined body we have loved so long. 

 

But where did New Haven come up with the mechanical design?  According to When Steam Railroads Electrified, by William D. Middleton, the New Haven bought its EP3 locomotives from General Electric, following a design developed for New York Central.  In 1929, the NYC ordered 22 electric locomotives to serve the new Cleveland Union Terminal,  taking passenger trains to and from the Cleveland Union Station, after steam locomotives were banned from the new station.  New York Central trains were powered by the electrics for 17 miles in Cleveland, from Collingwood on the east side to the downtown station, and on to Linndale on the west side.  General Electric and Alco built the 22 CUT electric boxcab locomotives, which were delivered in 1929 and 1930.  They operated in Cleveland until 1953, when diesels took over in Cleveland and the electrics were adapted for use in New York City.

 

Middleton says “The 80-foot, 204-ton boxcab locomotives were the first to use the articulated 2-C+C-2 wheel arrangement, which subsequently was adopted by the New Haven Railroad . . . and by the Pennsylvania Railroad for its incomparable GG1.”  (page 376) 

 

[richg1998]

From a simple Google search for prr gg1 frame.

http://www.steamlocomotive.com/GG1/quill.shtml

Search yourself. There is more. Just two 4-6-0's back to back. Not a big deal.

Rich

[aegrotatio]

It's not as simple as you think it looks.  I was looking for details on how it's articulated, since the GG1 has a rigid body.  Read earlier in the thread for those details.  Thanks anyway.

 

[daveklepper]

But the New Haven already had the 1-C-1+1-C-1 (2-6-2+2-6-2) EP-2 articulateds in operation, two Praries back-to-back instead of two Ten Wheelers, and the GG1 arrangement used earlier on the New Haven EP-3 could just have easly been a logical development from the EP-2.

[erikem]

Don't forget the Westinghouse motors built for the Milwaukee in 1919, which were two Pacifics's back to back. Even earlier were the GE locomotives built for the GN and B.A.&P. had articulated B+B wheel arrangements. The original Milwaukee electric's were 2-B+B. essentially an American articulated with a four wheel switcher.

The common element with the articulations was that the tractive forces were transmitted through the articulation between the truck frames rather than through the body.

{added note for readers unfamiliar with the details, Dave may know more than I do on the subject} The carbody of the GG1 and similar articulated electric and diesel locomotives is carried on the trucks in the way that a long rigid load would be carried on a pair of flatcars. There was a lot of engineering that went into the point of attachment between the carbody and trucks.

One advantage of articulation was that it reduced nosing (hunting) at high speeds, and that was further improved on the GG1 by the asymmetric wheel arrangement. Since the pulling force was carried through the articulation point, that force tended to straighten the trucks.

- Erik

[daveklepper]

This was also true of the EP-2  2-6-2+2-6-2, but was not true of the EP-1, which was a 2-4-0+0-4-2 or 1-B-B-1.   There, the "B"  or 4 wa pivoted like a regular truck, with the poney truck raidally pivoted from that truck.   The were orginalliy built a B-B's wihtout pilot trucks, which were added to reduce nosing on curves.   The EP-2 was built around 1914, if I remember correctly.    The EP-1 prototype was built in 1905.

[aegrotatio]

In what way is the wheel arrangement of the GG1 assymetric?

 

[erikem]

While the GG1's wheel arrangement for the whole locomotive was symmetric, the arrangement around the pivot points for each of the trucks was NOT symmetric. An FL-9 has an asymmetric wheel arrangement (B-A1A), but each of the trucks is symmetric around the trucks pivot points.

Hope this makes it a bit clearer.

- Erik

[cefinkjr]

aegrotatio

The G1 (half GG1) is one of my favorite shop hacks.  

When I described the G1 a while back, I was indignantly informed that this was some kind of urban legend.  Not so.  I saw this "urban legend" with my own eyes during a trip to the shop at Wilmington, DE a few weeks after the NYC-PRR merger that created PC.  The shop superintendent pointed out the G1 to me from a window in his office.  He told me that it had been rescued from an otherwise destroyed GG1 and was "off the books".

On a slight tangent, it always amazed me that a GG1 could ride like a Pullman while the Central's Class P-2b, with very similar mechanical specs, rode like every wheel had multiple flat spots and all of the springs had been removed.  (I was an NYC employee before the merger so there's no PRR bias in that comment.)

[erikem]

Chuck,

The GG1 had geared quill drive, which greatly reduced the unsprung weight, while the P-2b likely had nose suspended motors with a much higher unsprung weight than the GG1's.

- Erik

[Paul_D_North_Jr]

Someplace I have a few photos of that 'G1' from a 1968 +/- tour of that shop during an NMRA Mid-Eastern Region convention.  I believe it has also appeared in Trains - the old "Would you believe it ?" photo section - and a few books on the PRR or GG1's, etc. 

- Paul North. 

[Overmod]

In Trains with the comment "wouldn't two of these make a great engine?".  Prior to January 1966.

What remaining questions about the articulation are there at this point?

[BTG611]

So sorry to raise this post from the dead here, but I had the same question about how EXACTLY the parts all fit together.

I performed a google image search and found this, which really explained a lot for me, so I will leave it here as well for anyone else with this question to hopefully find.

You can see the "feet" mentioned prior as well as the central pivot point between the driving wheel truck castings, it's really a wonderful shot for explaining exactly this:

[BaltACD]

Paul_D_North_Jr

Someplace I have a few photos of that 'G1' from a 1968 +/- tour of that shop during an NMRA Mid-Eastern Region convention.  I believe it has also appeared in Trains - the old "Would you believe it ?" photo section - and a few books on the PRR or GG1's, etc. 

- Paul North. 

Over the years Paul D North's posts have always been intelligent and concise - with the following background - no wonder.  https://www.zatorlaw.com/category/paul-d-north-jr-p-e/

[CSSHEGEWISCH]

It looks like he puts as much effort into his hobby as he does into his profession, for which I am grateful.

[Pneudyne]

There is a good AIEE paper on the PRR GG1, which includes comparative lateral railhead force oscillograms for the P5a (2-C-2) and the GG1 and R1 (2-D-2) prototypes.  It is:

 

‘The Pennsylvania Railroad Class GG-1 Electric Locomotive’, by J.W. Horne (PRR) and H.S. Ogden (GE); AIEE Paper 60-48, 1960 January-February.

 

In respect of the running gear, less mentioned, at least in the railfan literature, are the stability aids that were incorporated to obtain satisfactory tangent-track tracking from the articulated running gear, the same also being used on the New Haven (NH) EP-4 and EF-3 classes.  These included a lateral cab restraint device and a rotational pilot truck restraint device.

 

I don’t have the specifics for the GG1 case, but those for the EF-3 are illustrative.  This is from the Railway Mechanical Engineer 1942 October:

 

Lateral cab restraint:

 

‘The cab -restraint device is located at the center of the cab directly above the articulation of the main frames. It consists of a double cam spring loaded roller device with a spring set up of 4,000 lb. for each spring.  It has been established from actual service that due to the accumulation of wear in the articulation parts, there is a tendency in an articulated locomotive to swivel the main trucks relative to the cab on tangent track when operating at high speeds, causing the locomotive to nose through the track clearance.  This may become critical if the natural frequency of the locomotive is synchronized with the nosing period.  The cab restraint device keeps the main trucks in line with the cab and eliminates this tendency to nose on tangent and relatively low degree curves, but permits freedom of movement without any restraint when operating on sharp curves. This device has given satisfactory performance for eight years.’

 

This device was developed by GE and described in patent US 2,111,428 of 1938 March 15 (filed 1936 February 11).  The locomotive type illustrated therein was the GG1.

 

Rotational pilot truck restraint:

 

‘A radius bar, spring restrained to 5,500 lb. is fitted to the inside of the truck frame to prevent oscillation of the truck around its center when operating at high speeds on tangent track, thus insuring smoother performance of the locomotive.  When operating around curves, the load on the restraint spring decreases.’

 

GE had several patents for this type of device, and I am not sure which applied to the GG1 and later 2-C+C-2 locomotives.  Initial use appears to go back the NYC T classes, which had B-B+B-B running gear, but not of the span bolster type. In the NYC case, the outer trucks acted as pilots to the inners, which rigidly fixed to their respective beams.  Also the pilot trucks had pivots that had no provision lateral motion, so that each assembly acted somewhat like a 4-4-0 steam locomotive.  Evidently it was found that rotational restraint, at least for initial movement, was required for tangent track stability.

 

Returning to the NH EF-3, the pilot trucks also had graduated lateral controls, described thus:

 

‘The four-wheel guiding trucks of the equalized type have rockers with variable restraint.  This restraint starts at 27½ per cent crack-off, increasing to 32½ per cent at 1½-in. truck swing, then dropping to 17 per cent of the center-pin load at full truck swing.’

 

This idea goes back to the Cleveland Union (CUT) terminal 2-C+C-2 locomotives.  The rationale for the variable restraint was given in the discussion part of another AIEE paper (1), as follows:

 

‘While in general the mechanical design of the locomotive follows conventional lines, there is one feature that is novel.  This is the characteristic of the gravity restraint device used for centering the guiding truck bolster.  This is mentioned very briefly in the paper.  In order to negotiate the sharp curves, a swing bolster is used with a 7 in. swing either way. The bolster is centered by rockers and specially shaped pads which rest and roll on the rockers.  The value of the centering force required to act against displacement from a central position is chosen to guide the front end of the locomotive on tangent track and amounts to about 35 per cent of the load on the bolster.  To avoid lurching of the locomotive when entering curves at speed, this force is increased to 42½ per cent at a displacement of 1¼ in.  With displacement beyond this value, the centering force decreases to a final value of 7½ per cent for a displacement of 7 in.  In using this type of characteristic, advantage is taken of the fact that the amount of guiding action required for a symmetrical articulated running gear on sharp curves is much less than necessary on straight track because when fully in the curve, the articulation guides the two main trucks.  The use of this characteristic gives desirable tracking qualities, especially by elimination of high flange wear on sharp curves.  This is of particular advantage in the switching service because the locomotives assigned to switching are on sharp curves a relatively large percentage of the time.’

 

 

Thus were heroic means required to obtain stable riding of bidirectional articulated truck type running gears at the highest speeds.  The full suite of “remedies” was apparently applied to the PRR GG1 and the NH EP-4 and EF-3 classes, but the two stability devices were not used on the CUT and NH EP-3 classes.

 

I imagine that the PRR DD2 2-B+B-2 prototype was fully equipped.  This looks as if it were an effort on the part of some PRR folks who perhaps wanted to resurrect the R1 2-D-2 type, but conceded the advantages of articulated trucks.

 

 

Incidentally, the CUT locomotives, although the first of the 2-C+C-2 type in the Americas, were not the first in the world.  The latter were a group supplied by English Electric in 1925 to the cape gauge Japanese Government Railway, later Japanese National Railway (JNR).  JNR not only had the first of this type, but also the last, in 1958, and the most, 237 out of an estimated 495 worldwide total.  They were nominally passenger locomotives, JNR using electrically similar 1-C+C-1 locomotives for lower speed freight work.  (There were also three nominally unidirectional 2-C+C-1 passenger prototypes, a wheel arrangement unique to JNR)

 

 

‘Operation of 3,000-Volt Locomotives on the Cleveland Union Terminals Electrification’, by F. H. Craton (GE) and H. W. Pinkerton (CUT); AIEE paper 33-36, 1933 January.

 

 

 

Cheers,

[Erik_Mag]

GE used a restraint system on the Bi-Polar locomotives built for the Milwaukee, my recollection is that the weights of the end cabs were used to provide the centering force. The articulation on these beasts were a bit more complicated than the 2-C+C-2 locomotives, with 12 driven axles spread across three frames. My impression was that the Bi-Polars did not like being run above 60MPH, with the engine crews on the Rocky Mountain district preferring the Westinghouse 2-C-1+1-C-2 locomotives.

The tests conducted by the PRR on lateral forces on track is what led the PRR to adopt the DD1. A caveat is that the DD1 had a high center of gravity for an electric locomotive which further reduces lateral forces due to an inverted pendulum effect. Similarly, the Milwaukee's Westinghouse motors had a higher center of gravity than the Bi-Polars and with the quill drive has a much lower unsprung weight.

Pneudyne's discussion reminds me of the chapter on articulateds in Withun's book American Steam Locomotives.

[Pneudyne]

The Milwaukee Bipolar was certainly odd, and perhaps unique.  Its running gear could be analysed in several ways.

 

Just one approach is to start with the assertion that basically, it comprised two permanently coupled, back-to-back rigid frame sets.  Each was essentially a 1B-D-0 unit.  Coupled, they became 1B-D-0+0-D-B1, conveniently shortened to 1B-D+D-B1.  In this case the '+' sign indicated a coupling between two frames, rather than say an articulated coupling between two trucks under a single frame, as in the 2-C+C-2 case.

 

Looking at each frame, the 'D' part was a rigid wheelbase, in the same way as for say a 2-D-2 locomotive, and not a truck as such.  The '1B' part was a leading or trailing truck.  As far as I know it was '1B', not '1-B', because the unpowered axle was part of the truck frame, in the same rigid wheelbase as the B part, albeit allowed some lateral movement but not rotational movement.  I think that the '1-B' designation would imply that the '1' part was a subtruck, and so allowed some rotational movement relative to the 'B' part.

 

The '1B' part was radial, connected to the frame at its inner end by a joint that was adjacent to the D wheelbase.  The couplers were mounted on the '1B' truck, so that all buff and drag forces went through the joint.  Here it gets arcane.  Did this make it an articulated joint?  I don’t know.  On the one hand, it certainly looks like one.  But on the other hand, the basic criterion for a single frame with articulated trucks is that all buff and drag forces were transferred between trucks (or span bolsters) by articulated joints, and in so doing, completely bypassed the main frame.  In this case the buff and drag forces did go through the main frame, as the D part was in the main frame.

 

As I understand it, lateral control of the 1B truck was through an inclined plane and roller mechanism, in concept at least broadly similar to what was sometimes used with regular pilot trucks.

 

On top of all of that, the superstructures of the two ‘1B-D-0’ units were shortened relative to the frames at their inner ends, allowing for the suspension between the two of a third cab, with no direct attachment to any of the wheelbase units.

 

Given that it two four axle rigid wheelbase sections, it seems unlikely that the Milwaukee Bipolar would have tracked and ridden as well as all-truck locomotives.

 

 

Cheers,