Chopped Sine Wave Question

I have not observed any of my locos, pre-war, post war, MPC, old (90s) or new, that "don't like" the chopped sine waves or whatever coming out of my Lionel and MTH variable power supplies.   

That said, I don't know how any of them, except one MTH I tested, run on a pure sine waves (it ran the same).  Maybe some of them run much better on older pure sine wave supplies.  But they all run on these new power supplies, and I'm happy enough with how they run.   

Lee,

I also ran all of my MTH and Williams engines on the chopped sine wave and I never observed any ill effects.  However, I believe that MTH recommended against the use of chopped sine wave with at least their initial release of DCS and others have reported problems with both MTH and Williams engines on power supplies with chopped sine waves.

Earl

I can't prove it, But my MTH Bump and Go Trolley and Hand Car, both run better on pure sine wave Atlas 80 and MTH brick running through the IR controller. By better I mean they seem to have the momentum, or power, or whatever to bump and go at much slower speeds than DC or chopped sine wave control. This allows them to play nice on the carpet, without moving the track each time they bump. And the trolley is now quiet enough to run above the TV all the time. So, even if it is magic, or my imagination, I'm going with it!

Also, I think my conventional engine speed control is a little better on the variable DCS output, than on the Z1000, with a Z controller, I used up until last year.

The problem with chopped sine wave is not the motor, but rather the electronic E-units. Some of them don't like it. Some vintages of MTH Proto One do not like to start/boot up with chopped wave power, but once thay are started/booted they are happy and run fine.

I do not know the technical reasons for this, but always suspected some poor/short sighted design engineering or maybe just malice on the part of the designer.

Originally posted by EIS:

"Lee,

I also ran all of my MTH and Williams engines on the chopped sine wave and I never observed any ill effects.  However, I believe that MTH recommended against the use of chopped sine wave with at least their initial release of DCS and others have reported problems with both MTH and Williams engines on power supplies with chopped sine waves.

Earl"

I should have mentioned that I run only conventional.  I imagine MTH's warning about chopped since waves and its not listing some Lionel power supplies as suitable for its trains has to do with concerns that its DCS syste might be confused as to signals by the shopped sine wave patters and the harmonics they wiould generatie in the track, etc. 

I think this situation arose with early QSI/MTH reverse units and sound systems. QSI introduced the first successful electronic reverse units (ACRU, then DCRU) around 1983. They were designed for Lionel transformers, which produced pure sine waves. They were designed for "Three-Rail Electrical Operating Standard (T.R.E.O.S.)" established by "Lionel's first class enginering department." When QSI sound systems came out, they didn't work well with newer transformers that produced chopped sine waves.

The original QSI/MTH sound system in MTH Premier Southern Ps-4 1401 (20-3006-1) that I purchased from a Forumite worked with my ZW but not with newer transformers. MTH Parts supplied a board and a chip from a RailKing Ps-4 (30-1125-1) made in 1997 and they worked fine with newer transformers.

According to what was mentioned in the video or lack of mentioning, there must be another issue with the Lionel CW-80 as it won't work with my MTH PS-2 engines, the Lionel person never mentioned if the CW-80 has differant electronics inside or not, just mentioned that it was a chopped sine-wave transformer. He claimed that the MTH Z-500 thru Z-1000 were chopped sign-wave, so in my opion the PS-2 engines should behave the same as when on the CW-80 as with the three mentioned MTH transformers. Post-war transformers were of the pure sign wave.

I am not an electrical genious or into electronics but the CW-80 deffinitly has some kind of electronic parts inside that the MTH transfomers don't have, as my PS-2 engines went crazy on the CW-80; all sounds starting up at once and forward and reverse constantly trying to be addressed within seconds of each other. Put the PS-2 engine back on an MTH transformer and no problems at all.

Lee F.

I was wondering if anyone has experienced any improvement in DCS signal strength with the Lionel chopped wave, especially with the newest ZW vs. Z4000

I do recall that my MTH engine #5405 (conventional engine with whistle, no battery) would not run at all on a Christmas layout a few years back.  I assumed it just had a problem and set it aside for repair.  I am very sure I was powering it with a CW-80 at the time.  I later put it on my permanent layout powered by a modern ZW and it ran fine.  I don't know whether the problem was due to a quirk in the engine or a problem with the chopped sine wave.  

My oscilloscope is out for an upgrade, but when it comes back I will make a dual trace comparing the CW-80 with the modern ZW and post the picture.

Earl

Just for my clarification would a pure sine wave transformer me old style like a ZW or a 1033?

What modern transformers are Pure vs Chopped?

Unless I'm missing something, if you are running in command-mode at full output voltage, then won't the wave-forms be the same for either type of transformer? The 'chopped' wave-form will only come into effect at reduced output levels as far as I can recall. So there should only be a difference for those running in conventional control.

I think you are correct Nicole.

I am by no means an expert on toy train transformers but I'm pretty sure that all PW type transformers are pure sine wave. All modern transformers are chopped.

The only modern exceptions are the MRC Pure Power Dual and the Atlas O/Williams 80 watt.

All other modern transformers use triacs to control voltage output to the track. I will assume that the AC wave form looks similar to this...

(ignore the SCR side, it does not apply)

(edit: I should clarify by saying that its the voltage controller that does the wave chopping. A Powerhouse 135/180 watt will be pure sine. A Z1000 "brick" will be pure sine. Its not until either of which is connected to a controller that the chopping occurs. Examples of controllers include: MTH Z controller, Lionel Powermaster, Lionel TPC 300/400, and I'm sure there are others too.)

What I understand from the video is that command control trains do better with the newer chopped wave transformers than the conventional, as it don't really matter which transformer for conventional usage. It has to do with the newer electronics charging a capicitor inside the new engines.

Chris, the older style ZW from around 1970 and earlier are pure sign wave, the newer ZW's are chopped sign wave.

Lee F.

Nicole,

the wave form on my modern ZW at full power still had a small amount of chop near the peaks.  I don't know if my ZW is typical or not.

Earl

Thanks Charlie, Flash, Earl & Lee, I was indeed missing something.   (If I only had a brain....)

Great explanation by Mike Reagan!  Helps me understand a lot about the new vs. old transformers and how they affect the newer engines.   Wish MTH did stuff like this.  

So what is the output on the variable DCS channels pure (smooth), or chopped?

The variable DCS output will be chopped. Only non electronic

power supplies like the old ZW, KW had variable sine wave

outputs using a moveable contact to change the output voltage.

The Z4000 had a variable output sine wave using electronics and

power amplifiers but that was the only one that did not chop the

AC to get a variable voltage.

The chopped output from most of the variable sources is not

very clean even at maxium output. The new ZW L comes the

closest to a sine wave at maximun output.

My thought on this is when using a chopped sine wave transformer coupled to a variable output on a TIU you get a whole lot more noise on the chopped sine waveform based on the 2 chopped waveforms combining. This is what causes the problems. The CW80 is probably a little dirtier than most and combined puts it over the top. I had trouble a few years back when I bought a Kline Allegheny and put it on a piece of track to test it with a CW80. When it powered up the sounds were scrambled. I put a caboose on the track with it and everything was then fine. A little load tented to smooth out the voltage spikes is my guess.

JMHO

Ron

The big difference I see between the two is the new transfomers like the Z-1000 will run all my trains at a much lower voltage and attain the same speeds as with an old transformer that will generally be 3 to 5 volts higher. An example is my 2025 runs well at 13.5 volts on an old post war transformer, the Z-1000 runs it well at only 10. Williams cruises at 8 volts with the Z 1000, using the post war transfomer it takes 13.

Rob

Earl, To answer your original question, it has to do with component selection and accuracy of tolerances.  The micro processors have to determing when high voltages are meant to equal a binary 1 or low voltages a binary 0.  Normally the processors are run on 5VDC.  So depending on the selection of the Capacitors and resistor that establish time to attain certain signal levels, the accuracy of the selected processor, the noise of the incoming power source all effect how the processor works.  A processor tolerances might be 3 to 5VDC is a 1, or it could be 4.5 to 5V is a one.  Same on the 0 side.   Voltages in the middle may be called unknown, preventing a processor from taking action.

Maybe an electronic transformer set a 10 Volts is sending just the peak and back half and the QSI early processors acknowledged the higher peak and determines voltage is above 10V and stay in the rest state even though the "Effective" Voltage was meant to be below 10Vs, or the noise of the converted AC causes the processor not to work properly and get more unknowns then definitive 1s and 0s when doing a calculation and it locks up.

I think many folks are jumping to wrong conclusions from that Lionel Video.

Lionel and MTH boards are designed from different models.  I can't speak to Legacy, but on the Lionel TMCC each component (R2LC and RS boards) take the AC power in and convert it to DC for use at each board.  For DC motor setups, the bridge rectifier is on the motor circuit board at the back end.

For MTH and older QSI the first thing that was done is the AC was taken in and rectified to DC, filtered to be a clean signal.  Everything on the output side is using DC.   That is why MTH PS-2 engines can run on DC.

Additionally, certain electronic transformer can chop the sine wave differently.  Lionel showed the front end chopped off and pulse at a high voltage for a short period of time to represent a certain voltage.  Other transformers may let a lower voltage sine wave form for a slightly longer period and cut the top off.  Both would have similiar effective voltages and power input to the device.  BUT, the device may respond differently to those wave forms.

AC Electronic tranformers have less maintenance, allow easier incorporation of Bells and Whistles signals, and may be less expensive to make especially when you want the ability to control it electronically from a remote (Like the moidern ZW, PM, TPC, etc...) but usually the output is noiser if needed to be converted to DC for electronic use. 

For MTH what ever that input AC looks like, the board changes it to DC and filters it,  Then the processors use it through PWM to drive lights, couplers. smoke units and motors.  What the motor sees on a MTH engine is controlled by the board and has less to do with the input wave form (to a degree).  For Lionel TMCC if they don't modify that AC until it get to the Motor driver board that can be a large reason why the motor will respond differently to different wave forms.  The DC Motor drive boards without cruise did not have a lot of filtering components on that board. 

I think people need to be careful of the conclusions they make from the Lionel Video.  It is a good explanation of how a AC wave form can/is electronically modified as an INPUT to a device.  If Lionels Technical Chief Engineer says their Legacy equipment runs better on their transformers modified wave form, I am sure it is so.  But that doesn't mean it is better for all types of train systems.  G

In my purely subjective experience, running both pullmor and can-motored conventional engines, both types favor a pure sine wave.  I like to run my trains slowly, and with a chopped sine wave (such as from the CW-80) both slow down much more on tight curves or grades, and then speed up on straights or downgrades.  The same engines will maintain a much more constant speed with a clean wave.  Sure, they smoke a little less and the headlight isn't as bright, but the performance is more important to me.

GGG,

Thanks for your very detailed explanation.  I think I now understand why some locks respond differently to the chopped sine wave.

Earl

GGG

Very nice explanation and I think you put it very correctly. Thank you for your input. I enjoyed the read.

For me, I have observed just the opposite.  With chopped sine wave transformers, all my open frame, atomic, and Lionel modern can motors run at a lower voltages and are nice and smooth even around turns compared to my post war ZW.  Smoke output is better too.

Again, just an observation, not scientific.

Interesting.  Are you using transformers that start at very low voltage?  For example, I typically run small engines at between 3 and 5 volts, and that produces the results I mentioned above.  Even the Lionel docksider exhibits the same behavior.

I don't think my ZW nor my CW-80 go below 6 volts, but I can make all my engines crawl using either.

I have only compared a dual set of CW-80s (one for the inner loop and one for the outer loop) G revision with my recently restored post war ZW which I still love and respect.

With the CW-80, I can make my 2025 crawl and still blow the whistle without it growling, all while puffing out nice smoke rings, and having a bright headlight.  With my ZW, I could never get the 2025 to crawl and still blow the whistle at the same time.  It needed more throttle to do both.

The only issue I have with the CW-80s is that at high voltages using a open frame locomotive like a 2025 and the air whistle it came with, is it will exceed the 5 amps the CW is rated for and the fold back circuit starts to kick in.  The 2025 runs too fast form my tastes anyway at these higher voltages so it's not an issue really.  It almost looks like its going to fly off the track on O36.

The ZW has no problem supplying all the current for the above and then some, but I get better low speed operation (not sure the exact voltage on the track) with better smoke and brighter headlights with the chopped wave CW-80.

Here is an o-scope of the CW80 made by someone else:

All "stock" CW-80s should throttle down to zero and run light trains down to one or two volts. But remember, due to the wave configuration, the output voltage cannot be measured accurately with a simple voltmeter. You can get some idea with a load on the device, but it will be of questionable accuracy. This is normal and proper operation. They are designed so that one can throttle back until the train stops with the lights, etc., remaining on, without tripping the e-unit, and then to proceed in the same direction 

KWs and 1033s, for example, can be configured to put out voltage down to zero, but you lose high voltage capability. ZWs are unable to put out less than seven or eight volts.

Except for the CW-80, we are talking strictly conventional operation, postwar models only.

If you look at the o-scope video above, the CW-80 does go to zero like wolverine49 says, I guess I don't have anything that runs on very low voltages.

Ironically, one of the defects in some of the very early CW-80s was that they wouldn't throttle down far enough to trip the E-unit. The remedy for this was to put a single light bulb on the layout somewhere to provide a little "resistive load," I think they termed it. The design calls for it to go near enough to zero to stop the motion of the train while keeping any lights on, and then moving the train in it's previous direction, without cycling the e-unit. This simulated a train pulling into a station, unloading/loading passangers, and then moving onward again. It takes a relatively steady hand on the throttle but becomes routine with practice.  It may help to know it is supposed to work that way. 

As far as I have been able to determine, the "G" models totally corrected this fault. 

I needed to replace the power supply for a freight yard and decided to purchase a Z-1000 for conventional control locomotives.

the PS-1 and QSI equipped engines either would not start up or ran very poorly using the Z-1000, which had replaced a postwar ZW-275w.

thinking it was the transformer, a 2nd Z-1000 power unit was borrowed and tested with the same results.

The  locomotives made in the late eighties/early ninties with QSI and PS1 boards may require pure sine wave Transformers.

BTW -The locomotives that had a big problem with the Z-1000 run just great using the Z-4000.

Okay did some of my on testing and this is not a certified test but just what I found on my layout with my equipment. This was all tested using Conventional mode only with handles on both transformers.

Lionel 0 gauge tubular track and layout is 13.5 feet by 22 feet Double main line. Flat board no inclines.

22 toggle switched blocks. 2 toggle switches for changing from Conventional to Legacy/TMCC

Fluke Digital Meter

Pure sine wave Z4000 MTH Transformer

Starting volts against a E7 AB Williams with True Blast II pulling 6 MTH Passenger cars.

AT 9.5volt no horn or bell at this voltage

At 10 volts got horn and bell but running at a good speed.

At 11 volts running at a good clip.

At 8.5 volts running at a correct speed but no horn or bell

At 7.9 volts engine stops

Now same setup with Lionel Modern ZW-C with two 180 bricks one for each handle.

At 7 volts light on and engine can blow horn and bell engine not moving

At 7.5 volts engine crawls and moves

At 9 volts engine running at correct speed

At 10 volts engine is high balling speed

Setup now is a Lionel SD60/TMCC with 9 freight cars.

MTH Z4000

At 7 volts nothing

At 8 volts some chatter in the engine

At 8.5 volts lights on dim and a clicking sound from engine still not moving

At 9 volts engine moves at a crawl but no sound or bell, smoke

At 10 volts engine running a correct speed with sound and bell with no smoke

At 11 volts engine running at a very fast speed but sound and bell and some sign of a little smoke.

At 12 volts engine running way to fast but smoke is working and sound.

Now same setup except Lionel Modern ZW-C with 180 bricks

AT 6.5 volts sound is on and bell no smoke engine now moving

At 7 volts engine starts to crawl with sound but no smoke

AT 8 volts engine a correct speed and sound and smoke working

At 9 volts engine high balling and sound and smoke working nicely

At 10 volts engine running to fast but all is working good

It seems to me for what I have tested for now the ZW-C wins the way for operations for me now. Seems the chopped sine wave makes the Lionel engines run much better than the pure sine. The Williams engine work much better at a crawl speed but did notice that the electronic E unit did not like the function to go into neutral with the handles and I had to use the direction button to get it to reverse but, not every time and that could have been the voltage at which I was at when I tried it. The direction button to me is not a deal breaker.

This is just what I found on my layout setup. Maybe this can help some understand the differences.

Thank you!  Thank you!  Thank you!

I love tests and data like this.  Very interesting.  Very well put down and organized.  

And quite interesting, too . . 

The shape of the chopped sine wave seems to me to have some of the same characteristics of pulse DC power that was popular for HO operation a few decades ago.  Both have a sudden burst of voltage and both make the trains perform better At low speed.

Earl

I have experienced very similar, if not the same, results on a Williams twin-motor diesel using a CW-80. Even my 1951 Lionel 2046 ('Old Growler') runs and smokes very nicely at lower voltages using chopped-wave power from either a CW-80 or a TPC.

What a great thread. (Thanks Earl!)

I was ready to replace my PW ZW with a new Z4000, now it looks as if I need to rethink that decision.

Thanks Lee

I found that the results for me was very interesting as well. I will try some more testing over the weekend to see if there are anymore things come up or not. This has become a very interesting thread for me. I have learned a lot so far and hope others have as well.