Bringing the AA20 (Andrey Andreev 20 ton per axle) back to life...4-14-4

[justme igor]

OK, so if I made a mistake in the 3 sentences above... tell me...

Better would be 4 meter 60 radii = 9 meter 20 diameter circle or more

being geared motors, they can get out of sync for sure.

The question is the gearboxes (assuming your siderods and crankpins and the wheels themselves can handle the torque).

so, with these gearboxes, if you turn the wheel, does the motor shaft turn easily? If not, I guarantee you will have problems, at the very least binding that will affect performance.

And the problem where i am not affriad of is: that the side rods can not handle the torck of the combined motors if one is out of speed on the other ones.

Best regards, Igor

 

[justme igor]

now you need more lateral travel in your drivers, or a hinged frame.

Will come.... if the prototypical wont work at all, first i am going to build a prototypical model, lets see what it takes to make it work, then a change in my track work/garden rail road---> before the steam one.
But the real one had more side clearance than you would expect, the front and back support had a lateral motion of 25 cm plus there was side play in the middle of the boiler where it rested on the frame of 15 cm.
on top of that the unflanged wheel could hover above the track, the 7 jong soviet engineers that delvelopt the aa20 toke everything out of the closet they could think of...exept...
The weakness of the early soviet track (splitting)(not even), couplers and they forgot totally about the turnouts....

 

[Greg Elmassian]

So, a bit of advice about "blind" drivers (no flanges).

As they leave the rail and then "climb back on" as would happen going into a curve and then coming out, derailments are often. Often models are made with blind drivers a smaller diameter so they actually "float" over the rail, and thus will not "catch" as they attempt to climb back on.

Also they can be (and often are) wider than the normal flanged drivers, but that cannot solve your issue as per your pictures.

Again, Aristo has solved this issue with hinged connecting rods, and large lateral motion in axles.

LGB solved with hinged frame, but I do not think that would be enough for your loco.

Clearly with the gearboxes you have chosen, you do not have the Aristo option.

Also, I think you are relying too much on the siderods handling the torque of the multiple motors. It's really not just the torque, when they slowly get out of sync, you will get binding, and the siderods will either break (and they will if they are prototype dimensions), or at the best case the binding will affect the power applied, and if you add lateral motion, that will make it worse.

But please proceed, I have been there and done that, as well as many of my friends. Sometimes you must learn by trying.

I would go metal siderods in any case, too bad you cannot purchase the Aristo parts, that would give you a big head start.

Best of luck, really!

Greg

 

[justme igor]

I just want to have a prototypical working aa20 in model, after that i will see what i must alter to get it to drive......
The garden layout can be changed easy for this one, with max 10 meter raddi curves if necessary.

Yes my blind wheels have 3mm side clearance and are 7mm wide/thick...

 

[justme igor]

So, a bit of advice about "blind" drivers (no flanges).

............

Anyway after a day of playing, i think i know where the hinges must be, surprise: almost prototypical and the pivoting points: surprise: almost prototypical
Also figured out some other needed measurements, yes it was a fun learning game
Copy and past anyone can do, but do you know what kind of knowledge there is behind it?!?!?!?
Keep on learning and dont be afraid to experiment and fail
Below every picture there is a explanation and a question or two.
Sorry if i named the things wrong, English is not my native language, and this is pretty technical English and not chit chat English




I will keep the 41 mm from back to back from the flange.


As you can see the clearance is just 2 mm at each wheel, to avoid problems the "bearing surface"of the wheels will go to 6mm each instead of 5 mm, the blind ones will go to 7mm(the flange is 1mm thick btw)
From the frog side to the rail side is 2.5mm in width inside
.


To let her drive over the frogs i will change the angle from 3 to 1.5 degree on the unflanged wheels, to help her on the frogs...will it help??? dont have a clue, we will see.
In the most desperate case i can make my curves a full mm wider to 46mm instead of 45mm



The aa20 in approximate dimensions(2cm to low btw, length is prototypical), in front just for the giggles the terrier loco. I could not resist to make this picture, sorry
Tray/drawer on the tender is just a idea to give a idea for the size.




Hinges of the tender are prototypical not in the middle, seems to work out very good.



The back support, THE link in between the drivering part and the feeding(Tender--->it was partially mechanical feeding the boiler not two fireman but one!) part, but the caring part.
As per prototypical there are some hinges that can not be done if you want to have it on a 4 meter radii curve.
Between the driver part and the bearing/supporting part there can be one hinge, i am thinking of a 2 hinged part between those two, the same for between the bearing part and the tender, that must be two.
If you look close at the drawing you can see that there is a two layer of steel that is on top of the bearing part.

Front support/boogie/bearing part, between those two i am in doubt one hinge or two parted hinge...
I think one hinge and the piviting point closer to the loco.

The body of the drivers is going to be much smaller/thinner 33 mm (was 35mm) to give more clearance and more play to the flanged wheels, it will gain 1 mm on the track without building/designing a track only for her.

In the red markings on the drawing you will see where the pivoting points will be in thick red, in thin red you will see the hinged parts.
The tips of the ball pens are pointing to the support that will slide over the driver body to give the much needed 8mm! side clearance.
Incl the 2mm extra clearance on the middle blind wheels and the extra 1mm clearance on both flanged wheels it could be possible to let her drive on a 4 meter radii curve and let her pass over some turnouts crossover etc (minimum 4.60 radii!!!) without trouble.
Even with all those adaptions she will still behave like her prototypical counterpart, especially on/with the pivoting points/side ways movement/ boiler bearing stuff.
I must admit i learn a lot more of things from the real life locos(/and tracks!+wheels), more than i would thought, my mechanical/engineering side was always strong, but never to old to learn.
Btw this is one of the many drawing i downloaded from the internet, this is just a guideline, only know width is the cab (3.08 meters) and the cylinders.

Thanks for reading, i hope you enjoy my endeavors, but she is tricky
With best regards Igor.

 

[Greg Elmassian]

I would give another piece of advice... your changing of wheel and track standards, that is very tricky. It's more complicated than you think, a lot of interaction.

I can see if you wanted to do something specialized to accomodate your loco, but then other rolling stock will not operate well.

Basically you can go the "toy trains" approach, which is smaller back to back, giving more "slop" between wheel gauge and track gauge.

Also then you have issues with turnouts, you have to increase the flangeway width, and really make flange-bearing frogs. Now the flangeways in your frogs are so wide, the wheels will drop into the throat of the frog. The only thing you can do there is increase the wheel tread width.

Basically turn back the clock on model trains trackwork about 50 years.

I wish you good luck, but would encourage you to follow G1MRA specs, and accommodate the curves with hinged modules with drivers, or more lateral movement in the driver axles or both.

Greg

 

[Paul M]

I must admit I'm enjoying this build description, it seems such a colossal project to take on. One thing that does concern me, is that as it's so big, heavy and long, are you planning on leaving on your track when it's complete? Because trying to lift it safely may well be a problem

 

[PhilP]

I must admit I'm enjoying this build description, it seems such a colossal project to take on. One thing that does concern me, is that as it's so big, heavy and long, are you planning on leaving on your track when it's complete? Because trying to lift it safely may well be a problem

The most sensible thing, would be to run into /onto a cradle / cassette, with a substantial pair of handles, and some way of stopping it rolling off when you carry it..

 

[justme igor]

The most sensible thing, would be to run into /onto a cradle / cassette, with a substantial pair of handles, and some way of stopping it rolling off when you carry it..

All the locos and rolling stock will be stored in a shed in the garden, the max train length is 6 meters times 8 tracks.
For maintenance or i hope not repairs, you are absolute correct with a "cradle" i saw some very nice ideas, incl "special transport" cases
But lucky I can drive them in and out of this shed.

Basically i wont leave the tracks
Best

 

[justme igor]

Sorry to report, but this project is on hold due to several reasons;
Starting up a second company(energy storage incl solar panels, incl support for EV cars, ect, send me a pm if you want to know more or order...EU ONLY!!!!) studying at the TU in delft, some new gained info on prototypical tracks and behaviour of the loco's, my house is still under construction atm( i am pretty behind with the goals i would like to achieve LAST year)....

Sorry, the project wont walk away, it is just delayed...
I hope you understand that a project like this, consumes a lot of time, withes i dont have the this luxerty for it right know.
Sorry again i wish it would have been different. I realy want to see the prototypical AA20 driving again next to my great grant fathers loco......
With best regards and stay safe all,
Igor

 

[Paul M]

Good luck with your endeavours Igor. I'll look forward to the next instalment, when you get started again

 

[justme igor]

Small update:
Yes there where some failures in between, i did not post them
Now this one is also no rose scent and pure moon shine, but i just wanted to see the thing drive.
Yes thing.....it is a long way from here to become a loco...
Wheels and axles are still not fixed, i just wanted to see her drive.
Some wheels need to be redone,, 3d printing is not always a hit...
It is still in experimental state!, but some dimensions are finally getting in the proper order...
With sorting out some things i am getting closer.
No i am not going to "middle" to get a good looking loco...

There is ALWAYS that one nail.......always.....


Motor compartment before i close it


Fire box suport(the small one) and the tender


I just wanted to see it drive....no comments wanted, it looks shit and it is shit, nothing is fixed, no wheel or axle or motor/wheel coupler is fixed, but it drove!


Thing is to big to take a normal photo....

My test track is holding up "good" but i need more room for expansion: 2.5 cm/1 inch per 10 meters.
With best regards Igor

 

[Greg Elmassian]

The obvious question is what is next? Testing the curve handling?

Just to follow, 4 out of the 7 axles will be driven?

What I am concerned about is the lateral movement needed in the drivers, but if all drivers are "blind" except the first and last, I think you will have to make the tread width much wider on the blind drivers. Lateral movement won't make any difference on those drivers since there will be no flange to "track" the rails.

Greg

 

[justme igor]

The obvious question is what is next? Testing the curve handling?

Yes i have some 4 meter, 5 and 6 meter curves, but first the wheels...
I am afraid that i must go for 8 meters....

Just to follow, 4 out of the 7 axles will be driven?

yes.

Page Not Found - Aliexpress.com

And they are strong....one motor can pull 6 kilo!!!!!!!if you pull up slowly and your loco is heavy enough!!! after that...it can do more, but i terminated the experiment, it still was eager to go...the frames of the cars (2 cars)started to bend with all those (dry)bricks....

I think you will have to make the tread width much wide

The gauge is 45mm, the flange to flange (outside) is 43mm atm.
i don't think i can go any wider than that on a 4 meter curve

What I am concerned about is the lateral movement needed in the drivers, but if all drivers are "blind" except the first and last, I think you will have to make the tread width much wider on the blind drivers. Lateral movement won't make any difference on those drivers since there will be no flange to "track" the rails.

The first two pair and the last two pair of wheels sets will have there corners, so yes 2 drivers on 4 flanged axles.
3 degree angle



The blind drivers had a angle of 1.5 degree.
It will be 0 and have the minimum od of 50mm instead of going up to 52.3mm(for the flanged wheels) 10mm thick total,


The green bar is 7mm high...

But this is the third set of wheels, the second set had also a degree in the flange itself.
So as you stated before the unflanged will be "hovering" above the track.
I would like it to see to touch the track with some weight, i need some weight also for traction, i dont want her to pull just 5 cars, but 50.

The unflanged wheels will be probably 51mm od with no degree
So i need to find the middle centre for in the curves...so much for prototypical.......F***
Yes i know this up forehand, but i had to give them a try...
Yes it failed surprise...
(and how much data is there left on the AA20?, not even the width of the working room, i really scraped the barrel empty, only three Russian newspapers left to be translated)
Yes the a17 tender you can get some measurements of for the loco
Yes i am (too) ambitious, i know, i build the frame already 3 times....to prototypical correct, it is a freaking challenge/nightmare

But when i am done i will have all the perfect dimensions to cast her in metal and make it live steam...your hear somebody singing? in the year 2525......

With best regards Igor

 

[Greg Elmassian]

If you make the tread width of the blind drivers wider, it has less opportunity to "drop" inside or outside the rail.

If you can make them significantly wider then they stay on the railhead for sharper curves.

You may also not taper the wheels, it really serves no purpose, as all you hope to do is increase friction between the loco and rails. This assumes you will bring the diameter of the wheels to allow them to touch.

Of course, you MUST have some type of suspension for the axles, or this will never work, The wheelbase is just too long, and even an exactly scaled down specification from prototype for track warp is not possible without a suspension for the wheels.

It will be a tricky build. Personally I would take the modular gearboxes from an Aristo loco and make your drivetrain.

You can add as many drivers as you want, and they already have the large lateral motion, and will handle track warp which is essential on a long wheelbase loco.

All drivers driven and all drivers flanged:

p.s. this loco runs on a 10 foot diameter / 5 foot radius curve. (still not sure if you are talking radius or diameter in your posts)

 

[justme igor]

(still not sure if you are talking radius or diameter in your posts)

8 meter of diameter is 4 meter radii.
I use radii: 4 meter, 5 meter and probably for the big ones 8 meter
In diameter it would be 8, 10 and for the big ones 16 meters, yes big circles.

Turnouts from:

1:9 = 6 meter radii --->sidings/yards
real life length 32 meters good for 40kmp/h, in model 90 centimetres

1:12 = 9 meter radii
real life length 38 meters, in model 1 meter, max rl speed 60 km/h

1:15 = 18.5 meter radii
real life 47 meter, in model 1 meter 40, max real life speed 80 km/h

1:34 - 99 meter - 140 km/h, in model 2 meters 15
But this is to fast for what i am after for...
got a bit of a dejavu feeling...i told you this two times before incl that i am using radii and not diameter

 

[Greg Elmassian]

Thanks, so the numbers you often use are radius.... so when I see 4 meter curve from now on I will assume 4 meter radius (radii is plural).

so you have 4 and 5 meter radius curves and you stated you may have to go to 8.

So just to calibrate my imperial mind... 8 meter radius is 26.25 feet, i.e. an entire circle is 52.5 feet. So a loop would take the entire width of your property? (trying to understand your situation).

In the USA it is frog number / diverging angle. Do the prototype railroads where you live express the turnouts in curvature? This is new to me.

Your frog numbers look like they make sense, in the hobby we think a #6 would be great on a main line, but you state it would be used in a yard, how true. Here in the US, it seems mainline are about 15 to 24, but I am on the old Santa Fe, and it's fairly high speed.

I was trying to look up standards for UK switches, and I saw what we call a frog is a "common crossing element", and they list a 1 in N number, which is what we call a frog number.

Sorry to derail your thread, but a link to the prototype standards for turnouts would be great if you have one.

Greg

 

[justme igor]

so the numbers you often use are radius.... so when I see 4 meter curve from now on I will assume 4 meter radius (radii is plural).

I only use radii

so you have 4 and 5 meter radius curves and you stated you may have to go to 8.

I stated several times that my "railway"will have 3 or 4 lines
2 lines with a min of 3 meter/9.84 foot----> next is 3 meter 15 cm/10,33 foot radii, would be 15 centimetres/5,9 inch centre to centre difference, next would be 3 meters 30 centimters/10,82 foot centre to centre
The third track will be only for the big ones would have a 4 meters/13,12 foot radii, but i was/am considering 5 meter/16,4 foot radii.(depends on the aa20)
But to get it more prototypical i am considering now for the "big ones track" 8 meter/26,24 foot radii, for 5 axles or more locomotives....those coal haulers.
(the wheels are going to be a big problem if i want to keep them in proto, 4th set for the aa20 will be printed this weekend....sigh, my bar is to high i must lower it...F@@@)

So a loop would take the entire width of your property? (trying to understand your situation).

The garden is(if you must know...) 60 x 100 meters/65.5 yard x 110 yards, for dutch standards it is pretty above average, for us of a it is small.....
We have a field for goats also....100 meters by 20 meters/110 yards by 22 yards, they will be out when i going to start constructing Amsterdam Central station from 1932(retirement project) pre catenary period.
My great grand father drove the 6300 series with the kettle of the 3900 (the beul/torturer,) and loved the aa20, i wonder where i got this from.....he would love to see or drive that thing on Amsterdam CS...
Nevermind the gauge difference!

In the USA it is frog number / diverging angle. Do the prototype railroads where you live express the turnouts in curvature? This is new to me.

I have no clue about frog numbers... i dont calculate anymore, i use programs for the drawings and it spits out the radii...
But if you want to calculate the frog numbers for me? the 5 way switch i just made has as smallest radii 9 meter 20 centimers/30,18 foot/362,20 inch...next in line is 13.20 meters....nok yourself out...there are 10 frogs....

I was trying to look up standards for UK switches, and I saw what we call a frog is a "common crossing element", and they list a 1 in N number, which is what we call a frog number.

I understand math, you would be surprised...It is far better then my Queen English or US of A or NZ or Australian English.(incl technical English in various fields)...
I dont think we have standards in the Netherlands anymore sinds 1930, but we have one of the most trains running per mile in the whole world, in a very small country...
Probably you will find some "standard" turnouts in France, Germany and in Poland, but i am not sure, regarding the EU.
(to give you several repeated answers on several repeated questions in one go...you understand this)

but a link to the prototype standards for turnouts would be great if you have one.

You can order some standard frogs, for sure, all maganesium alloy, suited for radii no number.

We have more fabrics like this:

Wissels voor elke spoorsituatie

Wissels zijn een belangrijk en specifiek onderdeel van het spoortraject. ETS Spoor is exclusief leverancier van Vossloh Switch Systems op de Nederlandse markt en biedt u naast het product ook...

www.etsspoor.nl

Incl Tata Steel
Have fun, i think google will translate

Sorry to derail your thread

As long it is not going to be a "mud fight" again or bickering again, I am ok with that, most of my threads (questions/show ect.) are pointing to one project and all is in one conjunction, one can not exist or work without a other....they need to join and come together....just like real life rail roads...."our hobby has everything but......" i will give you a hint, something with the wheels...
Best

 

[justme igor]

A update on my AA20 build:
After some experimenting i think i am there where i want to be.
The measurements, how it would react, wheel contact surface, bearing, flange to flange width, ect
Also figuring out how to make a "standard" for future builds.
The wheels can be adjusted from 42mm to 45mm

The frame is ridicules big, it wont drive on a 2 meter radii.

The overhang is 12mm/half a inch....


On a 3 meter radii:

With only all the drivers mounted and not the front runner/boogie or the back support it could run slowly on a 3 meter radii.
If the driving rods would have 1 mm side clearance.



On a 4 meter radii things are getting better and runs smoothly happy back and forward on my working bench.
Needles to say, it will not go over my switches, the wheel gauge is simply to small.
All the big ones will have there own big loop, i will try if i can fit 6 or even 8 meter radii in the garden so that the wheel gauge can be bigger--->44 mm is the wish
To get connection on the rest of the garden track they will need there own switches to run over.
(For the smaller ones to get on the big loop)

To give a idea on what is going on with the interaction between the rails and the wheel flanges:

This was on the 2 meter radii.

Now the 4 meter radii:



The front runners are in need of movement to be able to follow the track, probably i will make this articulated, to ease my pain a bit.


The back support is going to be articulated somewhere in the neighbourhood of the pencil.
The overhang is just to big.
The connection with the tender would be to risky also if i would leave it un-articulated.


Probably the tender will get the same type of "frame" as the loco.
This type of "frame" works pretty easy and can be assembled with "standard" materials.

Loco with tender

I also finally got around to do a bit on the electronics.
Again easy to install and cheap reclaimed (tested) old laptop cells.

I have in there 7.2v nominal with ~3 Ah, it should give me a driving time of a small hour, before the bms shuts off, if it is not enough, there is room for some more cells


With a charger i can just leave it working without worry's.
Plug and charge, unplug and play
Progress is slow, i am pretty occupied with some other more important things in life..
You can do a lot with a 3d printer, but for this kind of work, no, (Who am i kidding? i am a craftsman) to "dress up" the loco's yes it can work very good.
To make a object for metal casting, a 3d printer is a must have.

Getting high on my wish list: a lathe, cnc machine and a electric 3 kg furnace, the other (forging) furnace i have is way to big for this....

I hope you enjoyed the update,
With best regards Igor

 

[Greg Elmassian]

Interesting. One item was not making sense to me, although I am sure you noticed.

In the picture, you are measuring across a non-driven wheel, which can slide side to side:

But it makes little sense to put a straightedge here since you have omitted the 2 axles and drivers that CANNOT slide sideways, and would show you need way more than 1 mm more play in siderods.

Clearly that motor setup cannot fit on that track at all. So not sure what this picture shows at all, other than an unrealistic situation.

But I know you actually know that, just questioning why even show this with a caption about siderod clearance.

I think this has proven to you that unless you take the Aristo-Craft approach, your only hope is an articulated chassis.

I'm also a bit concerned about the thickness of the flanges, but as you say it will be on an outer loop for this loco only. They still might need to be thicker just for wear reasons.

One thing would be that you can put switches in that loop, and bring other locomotives "out" to the outermost loop, just the AA20 can never "escape" that outer loop. So even a switch with a 1 in 6 frog would probably be fine for your other locos, and the AA20 would always just go straight through, which should be do-able. You might have to play with the flangeway on the straight through stock rail, but that should not be a challenge.

Another crazy thought: what if you made the chassis a bit wider, and could angle the outer 2 gearboxes to stay more perpendicular to the track? Might be more trouble than articulating, so maybe not a great idea.

Greg