EMD F7 in SCALE

Where users can chronicle their builds. Start one thread and continue to add on to it.

Moderators: Harold_V, WJH, hwboivin3

User avatar
BigDumbDinosaur
Posts: 583
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: EMD F7 in SCALE

Post by BigDumbDinosaur » Sat Nov 07, 2015 1:42 am

Andrew Pugh wrote:
BigDumbDinosaur wrote: When a mass like the hydrodynamic coupling is spinning at several thousand RPM...
Wow, how high are you spinning that thing? :mrgreen:
Full throttle is 2400 RPM, which not-so-coincidentally is the RPM at which the V-twin develops maximum torque. That is well below the 3600 RPM that stock V-twins are governed to at full throttle. Idle speed is 1000 RPM. Each notch on the throttle above notch 1 increases engine speed by 200 RPM.
Looking forward to your next part of the series!
I'm a bit behind on posting. Been busy with real work. :?
Have you shared your name, or shall we continue to refer to you as BigDumbDinosaur?
BigDumbDinosaur is my nom de Internet. Abbreviating it to BDD will reduce wear and tear on your knuckles. :lol:
Science makes it known. Engineering makes it work.

User avatar
BigDumbDinosaur
Posts: 583
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

EMD F7 in SCALE

Post by BigDumbDinosaur » Sun Nov 08, 2015 4:35 am

EMD F7 in SCALE
————————————————
In one of my earlier posts, I noted that the real F7 didn't have a separate frame, relying instead on a truss inside the carbody to support the machinery and absorb the buff and draft loads. Such a design isn't very practical in a model, so my F-unit has a separate frame and the body primarily plays a cosmetic role. Also, as I earlier noted, the frame has a dropped center to position the prime mover (Briggs & Stratton V-twin) relatively low in order to maintain clearance with the body.

A lot of riding scale Diesels are built on ladder frames fabricated from rolled shapes like angle and C-channel. This is style of construction is practical because the prime mover is small enough to not cause clearance issues with the roof of the body. For example, the well-known Backyard Rails F7 has a ladder frame built from sections of angle welded to form a tee shape, with no change in height from end to end. This was feasible because the single cylinder Kohler engine that powers locomotive is small enough to clear the roof. This is not the case with the V-twin.

As a beginning exercise in designing a frame, I had to decide the structural material I would use. Building a dropped center frame from angle and/or C-channel would entail a lot of cutting and welding. The cutting part isn't bad, as I have a pretty substantial horizontal band saw that can make accurate cuts. The problem would come in jigging and welding the mess.

Maintaining alignment in a 6 foot long weldment built up from angle—the weakest of the rolled structural shapes—would be difficult. Also, the resulting frame would not have good structural and torsional rigidity and thus would be prone to sagging and twisting due to the concentration of weight in the center (recall that the power assembly weighs about 200 pounds).

An alternative would be C-channel, which is a better structural shape for this application. However, I would still be faced with jigging and welding a lot of parts. Maintaining planar and longitudinal alignment in a complicated weldment would be quite a challenge. That's just a fancy way of saying that the frame would likely pull and distort from all the welding and end up with a twist or bow, or both. Frames like that tend to provoke derailments, since the trucks are operating in different planes, causing equalization problems. Needless to say, I wasn't at all keen on struggling with that problem.

I gave this a lot of thought over a period of several weeks, going through a small mountain of paper in the process as I made and discarded sketch after sketch. In the end, I devised a frame design with only three structural members, requiring a minimal amount of welding. During the design process, I had decided that rather than try to fabricate the basic structural components myself I would get a properly equipped job shop to cut and form the required pieces. I could then fit them up in my shop and take care of the welding.

As the amount of welding would be relatively small, the welds' effect on planar and longitudinal alignment would be small as well. Plus with only three members to join, jigging would not be complicated. I should mention that among my various items in my shop is a large piece of 3/4 inch thick aluminum tool plate. This piece is rigid enough to act as an accurate base on which to jig parts for welding, and hence gets a lot of use. In the process of designing the frame, I also concocted a method of jigging the parts to produce and keep the necessary alignment while welding.

Two of the frame's members are formed into the shape of a shallow U-channel, which is a simple structural shape to fabricate. The center section of the frame resembles a tub (in fact, the filename for the drawing says "tub") and is where the power assembly is mounted. I devised a six-point mounting scheme for the power assembly, four bolts securing the base of the V-twin to the bottom of the tub and the two lower bolts that hold the coupling support to the PTO adapter tie rods passing through the tub's rear bulkhead.

As some pictures are worth many words, here are illustrations of the frame.
frame_assy.gif
Frame Assembly
frame_forechannel.gif
Frame Forechannel, 1 of 3 Structural Members
frame_midchannel.gif
Frame "Tub", 2 of 3 Structural Members
frame_aftchannel.gif
Frame Aftchannel, 3 of 3 Structural Members
The first picture is the frame assembly, which is the basic frame weldment with other parts attached (the long narrow bracket with elongated holes going across the frame is the support for the air compressor and one end of the main air reservoir). If you study this illustration you will see that the weldment consists of the forechannel (second picture), midchannel or tub (third picture) and the aftchannel (fourth picture). The cut out area at the left (forward) end of the aftchannel clears the power assembly's pump adapter, as well as the pump's mounting flange.

Some of the holes you see in the structural members are tooling holes used for alignment on the jig plate. I made an adapter that allows me to use a Black & Decker laser level as a sighting device to align the three sections on a common centerline. I'll put up a picture of that arrangement in a future post.

Once I was satisfied with the basic premise of my frame design, I constructed a mockup in my basement shop to verify my calculations. With everything verified and corrections made as necessary, I submitted the drawings to a local job shop. The parts were laser-cut from 1/4 inch hot rolled P&O steel and then formed in a large brake press. The combined weight of the three members after welding was approximately 135 pounds. As I earlier noted, weight is your friend in a locomotive, and a strong frame is a good place to have it.

Laser cutting is generally quite accurate, even when forming holes, as long as you understand that there is always a small amount of taper in the cuts. In the case of holes that have to be a specific size, I specify them undersize and then finish them as necessary to get the proper size. In thick sections, I like to use piloted counterbores for sizing larger holes.

Of the three structural members, the tub is the most complex, as it is formed on all four sizes and the forward bulkhead has two bends for clearance with other parts. The three holes cut in the front bulkhead are there to provide unrestricted airflow to the V-twin's cooling blower. The tub's rear bulkhead is an exact 90 degree angle to the tub's bottom, since that bulkhead is an attachment point for the rear of the power assembly. Below is a picture I took when I attached the aftchannel to the tub.
frame_rv.gif
Rear View of Partially Welded Frame
In this view you can clearly see the aforementioned cooling holes in the tub's forward bulkhead, as well as internal gussets in the tub and crossmembers in the aftchannel. There are also tapping plugs welded into the vertical walls of the aftchannel, they being used for attaching various items to the frame.

I'll put up some more pictures in my next post.
Science makes it known. Engineering makes it work.

User avatar
BigDumbDinosaur
Posts: 583
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

EMD F7 in SCALE

Post by BigDumbDinosaur » Sun Nov 08, 2015 9:46 pm

EMD F7 in SCALE
————————————————
As I noted in my original post, I generated a lot of drawings as this project progressed. One of the most important of these drawings is a "general arrangement" layout of the chassis with the power assembly installed.
rolling_chassis.gif
Rolling Chassis
The view is from the fireman's side of the unit. The clarity of the drawing isn't as good as it could be due the 1024 × 1024 pixel limit of image attachments.

This drawing is not "complete," in that I intentionally left out some details, such as piping. The main purpose of the general arrangement drawing is to fix the locations of all the big pieces so as to find and correct interference problems or possible assembly issues. The drawing should give you a pretty good idea of how the chassis is built and why some of the design decisions, such as locating all power assembly components off the PTO cover of the prime mover, were made.

Starting from the left can be seen the fuel pump (small red round object buried in the frame), fuel tank, battery, air compressor, power assembly and hydraulic reservoir. Partially obscured by the battery and air compressor are the main air reservoir and alternator.

The alternator and air compressor are driven off an auxiliary PTO shaft attached to the V-twin's flywheel—the shaft is a standard Briggs part. The alternator runs at 1.82 times crankshaft speed and will maintain all electrical loads plus charge the battery at idle speed (1000 RPM). The air compressor runs at 0.42 times crankshaft speed and is able to fully charge an empty main reservoir to 100 PSI in about three minutes at idle. The drive to these two accessories is via industrial timing belts with trapezoidal teeth. The air compressor pulleys are off-the-shelf Browning parts, while the alternator pulleys have been reworked for the application. The pulley on the air compressor proper has a solid hub and thus adds needed flywheel mass to reduce chordal vibration in the belt.
air_compressor.gif
Air Compressor Mounting & Drive
The particular air compressor I used is a Champion Pneumatic Model 1, which was the smallest single-stage compressor that I could find that had a cast iron crankcase (units with aluminum crankcases tend to develop cracks around the crankshaft bearing bosses when driven by an internal combustion engine). Other than rotating the cylinder head a quarter turn so the discharge port was pointing in the right direction no rework was done to it. I did mount the air intake filter in a different location, as its normal mounting would have interfered with the body.

Below is a picture of some of the pneumatic plumbing.
air_piping.gif
Pneumatic Piping
All pressure hose in the unit is braided stainless steel with a Teflon liner, using JIC fittings. Solid lines are DOM or seamless stainless steel, also with JIC fittings. The locomotive would be difficult to stop without functioning brakes, so I've made every effort to guarantee that compressed air is available at all times. Also, all propulsion control valves are pneumatically actuated—no air pressure means no propulsion.

As the air compressor is turning any time the prime mover is running, it is loaded and unloaded by a solenoid valve in response to main reservoir pressure changes, essentially working like the air compressor in the prototype. Unloading occurs at 100 PSI and loading occurs at 90 PSI, and each time the compressor is loaded a reservoir drain valve is cycled to get rid of condensate.
alternator.gif
Alternator Mounting & Drive
The alternator is an integral regulator 30 amp GM unit that has been modified to reduce its maximum output to 25 amperes and full charge voltage to 13.6, making it compatible with SLA type batteries. It is mounted "backwards" (pulley facing to the rear), since that was the only practical way to do it. As a result, it runs in the opposite direction from how it would run in an automobile, which has no effect on functionality. The alternator is supported by brackets that are part of the main air reservoir assembly. The reservoir has fore-and-aft adjustment to align the alternator pulley with its counterpart on the prime mover. The belt is tensioned by swinging the alternator, just as would be done in an automobile.

Speaking of the battery, I use a 35 ampere-hour SLA battery designed for use in a commercial grade uninterruptible power source. The nice thing about this type of battery is its high output relative to its physical size, as well as being spill-proof.

Going to the rear of the general arrangement drawing, the hydraulic reservoir you see is the second design. The first design was a reworked Northern Tool product and for a number of reasons, it wasn't entirely satisfactory, leading me to design and build a new one. Reservoirs of this type work best when they are tall and narrow, as such construction results in better heat dissipation, as well as reduced aeration of the oil. Incidentally and despite common assumptions, more oil doesn't make for a cooler running propulsion system. The available radiating surface, which in this case, is the reservoir itself plus the piping, is more important than the size of the oil charge.
reservoir_raw.gif
Hydraulic Reservoir before Painting
The body of the reservoir is formed from two sheets of 11 gauge, hot rolled P&O steel that were formed into the shap of a U. I farmed out the piece parts to the laser shop, as I don't have the equipment needed to form sheet metal. The standpipe for filling it is a piece of 1-1/2 inch muffler tubing and the return filter mounting bracket was formed from 7 gauge, hot rolled P&O steel. All seams are MIG welded and I tested the reservoir with low pressure air for leaks. The reservoir mounting brackets are made of 1/8 inch thick angle and bolt to the sides of the frame's aftchannel, making it easy to remove and install the assembly. The reservoir is vented through the filler cap.

The return filter is an off-the-shelf Zinga 10 micron part, mounted on a Zinga support. Filtration is critical in hydrostatic systems, as the very tight clearances in the pump, valves and motors can result in a major malfunction if junk starts circulating through the system. That's another way of saying that a lot of expensive stuff can be ruined in short order if the oil becomes contaminated.

The reservoir's suction port includes an in-tank strainer to keep debris out of the pump. Spur gear fluid pumps are the most efficient types but also the least tolerant of contaminants. Anything larger than about 100 microns that gets into the pump can result in it jamming. Suction and return piping is 3/4 inch trade size type K copper, soldered with industrial acid core solder. I used a relatively large pipe size to prevent any pressure differential from building up. This is especially important on the suction side, where pressure drop at the pump inlet can cause damaging cavitation.

That's all for now.
Science makes it known. Engineering makes it work.

Andrew Pugh
Posts: 179
Joined: Wed Jul 12, 2006 9:22 pm
Location: British Columbia, Canada

Re: EMD F7 in SCALE

Post by Andrew Pugh » Mon Nov 09, 2015 1:27 am

Wow BDD, this is great. Hands down the most interesting build I have read about.

I see that you appear to be exhausting in the prototypical locations on the roof of the unit. Many people seem to opt for and exhaust down low out the side, with dummy stacks.

Your post regarding the idle speed/max speed and speed steps prompted me to read the 567 development document http://utahrails.net/pdf/EMD_567_Histor ... t_1951.pdf The prototype used equal speed steps of 75rpm from 275 up to 800. I'm sure you already new...

Looking forward to learning more about the control system!

EDIT: And also the user interface...

EDIT EDIT: linky no workie
Last edited by Andrew Pugh on Mon Nov 09, 2015 9:17 pm, edited 1 time in total.

User avatar
BigDumbDinosaur
Posts: 583
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: EMD F7 in SCALE

Post by BigDumbDinosaur » Mon Nov 09, 2015 1:58 am

Andrew Pugh wrote:Wow BDD, this is great. Hands down the most interesting build I have read about.
Thanks!
I see that you appear to be exhausting in the prototypical locations on the roof of the unit. Many people seem to opt for and exhaust down low out the side, with dummy stacks.
The fumes and smoke rise anyhow, so I'm just giving them a head start. :lol: The best part is when you give her full throttle with a heavy train in tow and black smoke belches from the exhaust.
Your post regarding the idle speed/max speed and speed steps prompted me to read the 567 development document http://utahrails.net/pdf/EMD_567_Histor ... t_1951.pdf The prototype used equal speed steps of 75rpm from 275 up to 800. I'm sure you already new...

Looking forward to learning more about the control system!
That link is a 404—remove the period after .pdf.

One of the interesting things about naturally aspirated, two-stroke Diesel engines is that their power vs. RPM curve is relatively linear, as long as the engine is operated in a range where the volumetric efficiency of the blower is relatively constant. That this is so was the reason why the Ward Leonard elevator drive that was the basis of the early Diesel-electric propulsion systems worked so poorly. Double the engine RPM approximately doubles the available horsepower. However, the generator's output quadruples with the same RPM change, overloading the engine.

EMC's solution was the use of a load regulator—effectively a potentiometer with a logarithmic taper—to excite the generator according to the engine's power output. Hence it was possible to evenly step RPM.
EDIT: And also the user interface...
I'll get to it eventually.
Science makes it known. Engineering makes it work.

Pontiacguy1
Posts: 781
Joined: Thu Apr 26, 2012 10:15 am
Location: Tennessee, USA

Re: EMD F7 in SCALE

Post by Pontiacguy1 » Mon Nov 09, 2015 3:56 pm

I've done the exhaust-through-the-stack thing before. It was with a small 0-6-0 type diesel switcher. It's all cool until your eyes start watering from the fumes coming back in your face.

I did look cool, though. The 3.5 HP engine tended to burn just a little bit of oil, which put out a nice haze coming right up the stack. Again, it was all good until the engine would be working hard and your eyes would start watering.

User avatar
BigDumbDinosaur
Posts: 583
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: EMD F7 in SCALE

Post by BigDumbDinosaur » Mon Nov 09, 2015 4:39 pm

Pontiacguy1 wrote:I've done the exhaust-through-the-stack thing before. It was with a small 0-6-0 type diesel switcher. It's all cool until your eyes start watering from the fumes coming back in your face.
There's a little trickery in the exhaust system I designed that results in the gas discharge velocity being high enough to expel almost all of the fumes well above eyesight level. Those who have operated the locomotive have not complained about exhaust bothering their eyes.

I did a similar thing in a reverse-flow exhaust silencer I made for a friend's F7, which is powered by a Kohler one-lunger. The noise level is low but due to the way the silencer is designed, the gas velocity is accelerated on the way out, causing it to stay away from the engineer. In fact, it does a pretty good job of blowing debris away from the track. :D
silencer_front.gif
Exhaust Silencer Engine-Side View
silencer_rear.gif
Exhaust Silencer Discharge-Side View
silencer_parts.gif
Exhaust Silencer Parts
silencer_interior.gif
Exhaust Silencer Interior Construction
silencer_installed.gif
Exhaust Silencer Installed in BYR F7
Science makes it known. Engineering makes it work.

Andrew Pugh
Posts: 179
Joined: Wed Jul 12, 2006 9:22 pm
Location: British Columbia, Canada

Re: EMD F7 in SCALE

Post by Andrew Pugh » Mon Nov 09, 2015 11:57 pm

BigDumbDinosaur wrote:...

The fumes and smoke rise anyhow, so I'm just giving them a head start. :lol: The best part is when you give her full throttle with a heavy train in tow and black smoke belches from the exhaust.
That, is just perfect. Have you any video you can post of the locomotive in operation?
BigDumbDinosaur wrote: That link is a 404—remove the period after .pdf.
Fixed it. Thanks for that.
BigDumbDinosaur wrote: One of the interesting things about naturally aspirated, two-stroke Diesel engines is that their power vs. RPM curve is relatively linear, as long as the engine is operated in a range where the volumetric efficiency of the blower is relatively constant. ...
Ah yes...I imagine it would be...I noticed that the intake ports are closed many (15?) degrees prior to the exhaust valves closing, so I can't imagine there being any more or less than about 1atm of fresh air charge in the cylinder. They truly use the blower as just that...a blower...blowing the exhaust clear of the cylinder for the next cycle. The turbocharged versions must have altered intake port position or exhaust valve timing (or both, perhaps) in order to allow the turbo to compress the air in the cylinders before the intake ports closed. I would think the change is probably just in the exhaust valve timing, less 'custom' parts that way.
BigDumbDinosaur wrote: I'll get to it eventually.
I can't wait. :mrgreen:

Andrew Pugh
Posts: 179
Joined: Wed Jul 12, 2006 9:22 pm
Location: British Columbia, Canada

Re: EMD F7 in SCALE

Post by Andrew Pugh » Tue Nov 10, 2015 12:06 am

BigDumbDinosaur wrote: There's a little trickery in the exhaust system I designed that results in the gas discharge velocity being high enough to expel almost all of the fumes well above eyesight level. ...
You snuck in another post while I was writing mine...

I was doing some thinking recently about the smoke in the eyes issue with roof exhausts, and was thinking exactly along these lines...get the velocity up sufficiently to blow the exhaust above everyone's heads. There certainly is power to spare with the v-twin, a little restriction in the exhaust isn't going to hurt.

User avatar
BigDumbDinosaur
Posts: 583
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: EMD F7 in SCALE

Post by BigDumbDinosaur » Tue Nov 10, 2015 2:40 am

Andrew Pugh wrote:...a little restriction in the exhaust isn't going to hurt.
Except significant intentional restriction isn't necessary.

Exhaust gases have a lot of energy in the form of heat and exit from the cylinder at high velocity, producing the noise. There are two ways to dissipate that energy without making a lot of noise: restriction or expansion. Expansion is better but the silencer has to be quite large in order to for it to be effective. If expansion is confined to a series of chambers whose structure is such that the expansion shock waves are not conducted to the environment, the apparent noise level is reduced, but without significant restriction. Furthermore, the heat in the gases can be retained so that when they do finally exit to the environment they dissipate via convection. That is the principle upon which the above exhaust silencer works.

Below is an illustration of its innards.
exhaust_silencer.gif
Exhaust Silencer Interior Views
This is what is referred to as a reverse flow silencer, since the exhaust gases travel to the rear of the unit and then start expanding toward the front (front is on the right). There are four expansion chambers, the largest one's volume being approximately 10 times the engine's displacement. The three baffles that separate the chambers are pierced with 143 equal-size holes. The hole size was calculated to produce enough pressure differential to promote expansion, but not enough to generate significant back pressure. The total area of the holes is approximately 1-1/2 times the area of the tubing that acts as inlet and outlet.

Here's what the baffle looks like.
exhaust_baffle.gif
Exhaust Silencer Baffle
About half of the total expansion takes place in the first and second chambers. The third and fourth chambers are responsible for killing the high frequency noise component, which is what is usually objectionable to most people. By the time the gases exit the silencer their pressure is only a couple of atmospheres, so the noise level is low and it's all low frequencies.

In order to prevent the silencer from itself acting as a resonator it is constructed from 3/16 inch wall rectangular tubing. The baffles are plug-welded top and bottom so they do not resonate—also, the welds keep them from drifting out of position. The inlet and outlet tubing is 1-1/2 inch exhaust pipe, nothing special. In order to retain as much heat in the exhaust gases as possible, the silencer is wrapped in an insulating blanket, which also keeps it from heating up the interior of the carbody.

This last picture is of the exhaust pipe, which exits out of the engineer's side of the loco down by the false fuel tank.
tailpipe.gif
Exhaust Pipe
The guy for whom I made this thing is quite happy with the exhaust tone. The mechanical noises from the engine almost drown out the exhaust, which is a low rumble. As a bonus, the locomotive got heavier. :lol:
Science makes it known. Engineering makes it work.

User avatar
BigDumbDinosaur
Posts: 583
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: EMD F7 in SCALE

Post by BigDumbDinosaur » Tue Nov 10, 2015 3:01 am

Andrew Pugh wrote:Have you any video you can post of the locomotive in operation?
I don't have any, but a while ago someone referred me to a video that was shot in 2011 that I didn't know existed. The F-unit, which was running sans body (it had yet to be built), was pulling a bunch of cars. The first scene is at 00:28 and there is a second one at 02:52. There's a brief third one at 04:20, in which the train leaves the station with a bunch of passengers. You can't see the engine in that one, but you can see the full train length for a few seconds.
Science makes it known. Engineering makes it work.

Andrew Pugh
Posts: 179
Joined: Wed Jul 12, 2006 9:22 pm
Location: British Columbia, Canada

Re: EMD F7 in SCALE

Post by Andrew Pugh » Fri Nov 13, 2015 1:26 am

Your locomotive pulls away nice and smooth, no jerking there. I hope to see more some day. The puff of smoke at throttle changes under load has me particularly interested. :mrgreen:


Now, this is off topic, but I spent several days on and off searching for this data, so I'm including it here...since I already mentioned it for the 567...I hope you don't mind BDD.

EMD 645 locomotive diesel engine RPM vs throttle position:

Low idle: (I saw this figure posted in several places while searching for rpm by notch, of course now I can't find it...but will post it when I do)
Idle: 307
N1: 307
N2: 360
N3: 490
N4: 546
N5: 627
N6: 728
N7: 812
N8: 900

These may not jive with published technical specs (possibly contained in maintenance literature I could not access for free from the internet...), rather these were observed speeds noted during emissions testing of an SD40.

Not the nice even 75rpm steps for the 567 from 275 to 800.

Post Reply