When was the change from "jerky" Lionel Odyssey to smooth Odyssey motors?

Can’t help with a date but the change was to the DCDS board that drive the motors, not the motors themselves.

Pete

I only had 1 engine that exhibited this. A NH Mountain. Was delivered around 2001. I purchased a NYC 0-8-0 and a 10 Wheeler that were fine the next year.

In my head, that when you jump from 6-28xx series in steam to later 6-38xx series in steam- but that's probably not the exact correct line in the sand.

That said, I download and read the manuals and examine the pictures on ones I'm considering buying to see if it has the physical Odyssey ON/OFF switch.  I believe the switch as added was the same time the firmware code and hardware improved.

While at it, since that may be where I got the manual is from Lionel support/parts, I'm also seeing what parts are available or more specifically not available to help in my purchase decisions.

I have some of both and yes, very much now recognize the Odyssey speed jump in my 6-28078 Odyssey Pennsylvania 2-10-4 Texas as an example.

The only people who can answer this work or did work at Lionel developing or servicing these boards. It was a topic at Lionel service school but don’t recall that even there, a specific date or part number(s) were given. The change was in the algorithm in the firmware.

Pete

FWIW, the Odyssey "lurch" was only on initial power-on, starting and stopping under power didn't have the issue.  I do believe that the later TMCC Odyssey I boards also had smoother running, but maybe that's just my imagination.

If you have Legacy the lurch will not be present since Legacy uses true speed steps when you ramp up the throttle.  TMCC  jumps two-three steps when you turn the red knob on the cab1.  Hence the first cruise control boards would lurch at start up (firmware issues).  I think you could get updated boards at the time but not sure.

I had a JLC challenger once and sold it due to the lurch and other issues I didn't like.  Stupid me, If I had known then what I know now, I would of kept it and had @Bruk convert it to Legacy.  Still kicking myself over that one

The infamous Odyssey lurch had nothing to do with any of the motors of the time. There was an open-loop uncontrolled power application to the motor(s) possible at first track power-up and throttle application with the earliest boards. If track power was not cycled, the board would function normally during the course of running. Some engines like my RS-11s at the time would launch themselves over turnout power drops momentarily before regaining control. Lionel did eventually fix the issue and would send you new boards if you complained. My JLC Challenger had the problem too. I'm pretty sure the code was modified but I'm not sure if the hardware was.

There were at least two distinct speed curves before going to on/off Odyssey sometime in 2002-3 (can't remember).

The Odyssey speed curves were never quite optimized for driver size, gear ratio, and max motor RPM. Here's a plot of my PRR S1 with the on/off Odyssey board. I do not know what firmware version they were using but most of the DCDS boards would have a small sticker that probably ID'ed it. You may notice the system only had 20 usable steps before the motor speed maxed out.

Nice chart Norm, it's interesting that Odyssey I seems to only have around 25 speed steps before reaching warp speed.

I don't know what engine could have used the full set of early Odyssey steps. Maybe something with a 10:1 gear ratio?

I want to say that the Odyssey Hall Effect rings back then could have had different resolutions too (number of magnetic pulses/rev).

Another thing is the application of speed commands from the CAB-1. The Big Red Knob behaved like a 'velocity pot' and could issue speed commands at different rates depending on how fast one spun the big red knob. Spinning it slowly would dole out steps one at a time and cranking it like a garden hose spigot would issue them forth at 5 steps per blip. Not sure how many people know or remember that.

I've never seen an Odyssey I ring with anything but four magnets.  That doesn't mean there never was one, just that I never ran across it.  I've replaced a lot of magnet rings over the years, I keep a few in the parts box as they are a regular repair item.

Well my S1 has 12 pulses per rev. What kind of motor had 4? Maybe that was Lionel's rough way of adjusting speed curves. It was said some of the first Odyssey engines only had two pulses per rev then increased later on. I think the very first Odyssey engines would have been the NYC Mikado, the PRR PAs, and then of course the Odyssey demonstrators.

https://ogrforum.com/...odyssey-demonstrator

To make things fuzzier, I have 2 USRA 2-6-6-2 Mallets, both from the same original run of these locos. Same item number, so same parts, one would think.

But - one lurches, one does not. One I have nicknamed "Lurch", for obvious reasons. Both run very well - once Lurch has gotten past his startup spasm.

I have a few lurchers; I don't like it, but since it is one and done during an operating period, I just try to think of it as part of the Startup Procedure, along with picking up the CAB-1 and powering up the track.

Norm, does your S1 REALLY jump from zero to 6.6 scale mph at speed step 1?  Wow, hang onto your coffee cups in the dining car!

so run that by me again gary. the odyssey I lurch does not happen when using the legacy cab2. is this for all odyssey I engines. if so then why? is it the 200 legacy speed steps?

I think the change occurred sometime in late 2001 or 2002.

The jerking only occurred when operating conventionally.

@NYC 428 posted:

The jerking only occurred when operating conventionally.

Wrong!  The initial power-on startup lurch most certainly did happen in command mode!

@Norm Charbonneau posted:

Well my S1 has 12 pulses per rev.

I'm confused by that statement.  Are you saying the magnet ring had more than four magnets?  The part for the TMCC S1 is the same as every other Odyssey I locomotive I looked up, and it's the same part number as I order for replacements.  I know for 100% certainty that has four magnets in the ring.  It's the same part number I've always ordered, and there are a few in my cart for my next order.  It's also the only magnet ring I find on the parts site.

The old S1's magnet ring had twelve 5V pulses when rotated by hand and measured at the hall effect pickup. Cross checking with the frequency counter and tach verified. If anyone wants to go down the rabbit hole, here's a little vid about it:

@Bob posted:

Norm, does your S1 REALLY jump from zero to 6.6 scale mph at speed step 1?  Wow, hang onto your coffee cups in the dining car!

It steps out smartly!

@Lionelzwl2012 posted:

so run that by me again gary. the odyssey I lurch does not happen when using the legacy cab2. is this for all odyssey I engines. if so then why? is it the 200 legacy speed steps?

I have found I can minimize the lurch using a Cab1 by watching the command base and turning the knob until I see 1 flash on the Command base. Then wait for it to move. Part of the lurch is due to turning the red knob until you see it movement but by then you may have added 5-7 speed steps and it jumps out due to its delay in responding.

Using a Cab2 and moving the dial to speed step 1 does the same thing. Once it starts moving then increase the speed.

A more economical way to smooth the starts is replace the DCDS with a Cruise M.

Pete

@Norm Charbonneau posted:

The old S1's magnet ring had twelve 5V pulses when rotated by hand and measured at the hall effect pickup. Cross checking with the frequency counter and tach verified. If anyone wants to go down the rabbit hole, here's a little vid about it:

Well, I just looked, and it appears that the stock 4 magnet actually does produce 12 pulses for a rev, I just checked on my test stand.  I wouldn't have expected that, but that's what the 'scope says.  I had to back into it and compute from 3.4 RPM, and a 25ms period between output tach pulses,

So, I'm right, the magnet ring on this motor most certainly has only four magnet dimples.

And you're right, as it sure does produce 12 pulses, something I sure didn't expect!

Rabbits are safe.

@gunrunnerjohn posted:

So, I'm right, the magnet ring on this motor most certainly has only four magnet dimples.

And you're right, as it sure does produce 12 pulses, something I sure didn't expect!

Wha.. ?

How does that work? I could see there being eight pulses, 2 for each magnet, magnets having two poles. But how does it make 12 pulses?

Does the Hall effect sensor pulse on both N, S, and “off” maybe?

@rplst8 posted:

Wha.. ?

How does that work? I could see there being eight pulses, 2 for each magnet, magnets having two poles. But how does it make 12 pulses?

Does the Hall effect sensor pulse on both N, S, and “off” maybe?

Think of the magnet as “on” …..before and after is “off” so it’s off until it sees the magnet, then off as it passes it again…….so off, on, off,…X4 …

Pat

Pat, that would be 16 pulses for the four magnets.   I'd have to do more analysis to see what they're doing, but the output is certainly 12 pulses for each rev of the flywheel.  Perhaps they have an East pole on these magnets.

@gunrunnerjohn posted:

Pat, that would be 16 pulses for the four magnets.   I'd have to do more analysis to see what they're doing, but the output is certainly 12 pulses for each rev of the flywheel.  Perhaps they have an East pole on these magnets.

Off, on, off is 3 pulses, ….4 sets of 3 pulses is 12 …..automobile mfr.’s have been using the same Hall effect principle for decades…..although mostly opposite with a fixed magnet, and metal stator ….Ford used to use a TFI ( thick film ignition) with 8 metal blades ( V8 engine ) and 8 open slots 16 pulses were used by the module….8 to drive spark, 8 to drive injector pulse ..

Pat

@gunrunnerjohn posted:

Pat, that would be 16 pulses for the four magnets.   I'd have to do more analysis to see what they're doing, but the output is certainly 12 pulses for each rev of the flywheel.  Perhaps they have an East pole on these magnets.

The poles should do the separating, I’d think,….Ford always used off/on…that’s where a lot of techs would get confused and not understand how the system worked, ….the open blade ( no metal ) was also a trigger,…..,same as metal in front of the magnet…

Pat

The op asked about odyssey motors I thought Lionel never made the odyssey motor. Is this wrong did Lionel make locomotives with an odyssey motor and if they did how many did they make

Lionel made a handful of the Odyssey motor prototypes, but they were never used in a product.  Only a guess, but I suspect that the costs as well as the compact nature of the can motor and the ease of adding speed control to can motors killed the Odyssey motor project.

I can't find the exact sensor IC they are using (I have one with '2005' and another with '1021' on it). I guess its a good thing the old Odyssey system wasn't counting magnets and not pulses! Maybe the IC triggers for a rising, a peak, and a falling signal if there are only 4 actual magnets with 3 pulses per pass.

I'm sure there are only four magnets, I put the ring on the motor I'm using on the test set myself.  Obviously, someone that was doing the design knew exactly how many pulses were generated.  I would have expected either four or eight pulses, I was surprised to see twelve.  The 'scope doesn't lie, they were really there.   It would be interesting to know exactly what sensor they used and have a peek at the data sheet.

I don't have many pure Odyssey engines left as most have been converted to ERR. The S1 may get ERR one day if I keep it long enough. It got me to thinking though - I suppose I could print my own magnet ring and offset the speed curves by changing the number of magnets. Hmm.

@harmonyards posted:

Off, on, off is 3 pulses, ….4 sets of 3 pulses is 12 …..automobile mfr.’s have been using the same Hall effect principle for decades…..although mostly opposite with a fixed magnet, and metal stator ….Ford used to use a TFI ( thick film ignition) with 8 metal blades ( V8 engine ) and 8 open slots 16 pulses were used by the module….8 to drive spark, 8 to drive injector pulse ..

Pat

I hope Lionel used better heat sinking than Ford on the ignition module. 😉

@Bob posted:

Norm, does your S1 REALLY jump from zero to 6.6 scale mph at speed step 1?  Wow, hang onto your coffee cups in the dining car!

Most Odyssey steam starts around 5-6 mph on speed step one. I've got two locomotives with Odyssey (JLC UP Challenger and a SP AC9) and both run about 5-6 scale MPH on speed step 1. I verified by pacing both with a PS3 ATSF Hudson.

I've actually eliminated most of the lurch by running the Odyssey stuff with the DCS remote. I click the thumb wheel to speed step one and wait for the locomotive to start. Both locomotives will be getting a CCM in the future though.

If you guys are seeing 12 pulses, ( which I don’t doubt ) then there must be 4 segments with 3 pulses each, ….so the ring must be divided in 4 pie slices,….at least that’s how my mind sees it…..😉

Pat

@Norm Charbonneau posted:

I don't have many pure Odyssey engines left as most have been converted to ERR. The S1 may get ERR one day if I keep it long enough. It got me to thinking though - I suppose I could print my own magnet ring and offset the speed curves by changing the number of magnets. Hmm.

How about an easier way?   You can use my Chuff-Generator and set it to have as many or as few pulses for a rev as you like.  I've used it to replace the TAS tach reader board and to generate the duty cycle signal for diesel RS4 board prime mover ramp up/down signals.

@harmonyards posted:

If you guys are seeing 12 pulses, ( which I don’t doubt ) then there must be 4 segments with 3 pulses each, ….so the ring must be divided in 4 pie slices,….at least that’s how my mind sees it…..😉

Pat

That's what I'm seeing Pat.  I wasn't expecting three pulses for each magnet, but that's what I get.

Easier way? You guys sure know how to turn a molehill into a mountain. You must be really bored.

Pete

@Norton posted:

Easier way? You guys sure know how to turn a molehill into a mountain. You must be really bored.

Pete

LOL.

If there is a 3-pole armature rotating in a ring with 4 magnets, wouldn't that make it possible to generate 12 pulses per rev?

MELGAR

@MELGAR posted:

If there is a 3-pole armature rotating in a ring with 4 magnets, wouldn't that make it possible to generate 12 pulses per rev?

MELGAR

The Hall effect sensor has to have a magnet pass pretty close to it, …….beings the poles of an armature are inside the motor, the sensor in no way can read the poles,…..now, in the case of a Cruise M, it’s actually reading the the gaps in the armature’s commutator segments, ….but that’s a topic for a different discussion,…

Pat