Skip to main content

Well guys, while it would be really cool to have a scope, I worry that it would be a case of "pearls before swine". I can buy it. I could hook it up. I could get a picture, but I wouldn't know what it means. I want it to be a tool and not a toy. Without some serious education, I'm afraid that is all it would be.

I guess the $64,000 question is, what useful information could be gathered for this project, that only a scope could ascertain? Is it just something that would be nice to know, or will it really allow us to pinpoint some actual issues with the signal?

I suppose, I always have you guys to help interpret the test results, if I get it. I would need you to tell me what tests to run and what we were looking for. I'm game for that if you are. I would certainly learn as we go, so there is that. Before I spend the money, I just want to be sure that it's the right decision.

Thanks!

gunrunnerjohn posted:

I'd hold off for a spell.  While it would probably be useful to have one, I do question if you'd get your money's worth from it.  I principally use mine on the bench for development work. 

I think a better investment might be the track signal engine with a remote readout or audio tone to do some signal strength mapping.

Thank you John! This has been a fun and interesting side discussion, but it is good to see that we've come to our senses.

Let me go find my donor R2LC. A few years back, I had an early TMCC engine split a switch and fried the insulation on the internal wiring. I hope that board is still good, but I think it is. Did I see that project needed a 51K resistor and a .1μ cap? I can't remember, resistors in series you add? if so, I can make 51K or very close from what I have on hand. The cap, I have a drawer full. Times like this when you miss Radio Shack, though my luck they wouldn't have had it anyway.

Last edited by Big_Boy_4005

Thinking along the lines of a test engine. I wonder if that little scope I have could be used onboard?

Now I have not used it very much so I'd have to read through the manual. But I remember it has a record function. It might be able to work as a onboard data logger. So could you produce a signal map for a particular track route from the data?? 

Nick

GRJ,

Thinking about options to send data from a moving engine to remote location.  Am I understanding correctly that the "data" is a DC voltage from the R2LC board?  What is the range of voltages here?  As is often the case I'm thinking an Arduino could be used with one of it's A to D pins, then it could forward that data over a wireless module to another Arduino.  The basic example for the nRF24 library does just this without any, or much modification needed.  

JGL

John, that's exactly what I was thinking.  The voltage is from 0 to 5V, a good match for the A/D.  Something like the Arduino Nano A/D could capture the value and send it over a wireless link.  I haven't looked at the exact software, but I suspect a similar library would be useful.  I found this when I searched...

This NRF24 library has now been superceded by the RadioHead library http://www.airspayce.com/mikem/arduino/RadioHead RadioHead and its RH_NRF24 driver provides all the features supported by NRF24, and much more besides, including Reliable Datagrams, Addressing, Routing and Meshes. All the platforms that NRF24 supported are also supported by RadioHead.

I guess that would be the one to look at.

Last edited by gunrunnerjohn
Big_Boy_4005 posted:

Did I see that project needed a 51K resistor and a .1μ cap? I can't remember, resistors in series you add?

Correct - 51k and 0.1uF.  Anything near 51K will work just fine.  You might have 47k for example.  Just make a note of what you used when reporting results. 

Yes, resistors in series add.  Or, equal resistors in parallel divide by 2 - you may have 2 100k resistors which in parallel would make a 50k resistor.

I'm curious to see what GRJ and JGL come up with for a wireless gadget...but in the meantime for 10 cents in components and some careful soldering, you can make a signal-strength tool.   I say 10 cents because the multi-meter shown earlier is a Harbor Freight "freebie" meter which works fine for this application.  I saw a Harbor Freight free-with-coupon in today's Sunday paper for this exact meter...and snooping on your zip code I see there's a Harbor Freight nearby!  It comes WITH battery and for that price you can snip the meter leads to make them shorter and maybe drag it around on a flat-car.

 

cjack posted:

Wireless alarms would be up in the gigahertz frequencies. 2.6 to maybe 6.0 Gigahertz. It could interfere with your CAB2 to base. But not with your base to track for that frequency is about .000455 gigahertz.

Thanks Chuck, guess we can rule that out. I haven't used the Cab-2 yet, I'll get to that later.

I just got a chance to watch GLA 11, and Bob Bartizek was saying he only uses TMCC and not Legacy, partly because the Cab-2 didn't have the range he needed, but mainly for operations, he didn't need the "fluff". I concur, but I bought with one engine in mind, my Milwaukee road S-3. I'll reach for the Cab-2 when that baby finally hits the rails.

Big_Boy_4005 posted:

I have plenty of 47K's and 2200's. What's 400Ω between friends at that level?

It's up to you, but using just a single 47k (since you have it) will give you just as much information as trying to get it closer to 51k.  Just keep a note that that's the value used and you're good to go.  After all, 400 ohms is about 1% of 51k and I'm guessing you don't have a stock of anything better than 1% resistors anyway! 

Big_Boy_4005 posted:
 

An idea just popped into my head. It might be something, it might be nothing, probably nothing.

What part of the radio spectrum is used for wireless alarm communication?

Earlier I think you said you were getting an old-school AM radio to sniff around the layout.  Note that with R2LC meter output you are building a sniffer for signals around 455 kHz.  So in practice what you should do is run the train with meter around the layout with the TMCC base turned OFF to get a background/baseline reading.  The meter will NOT read 0 and will probably vary around the track.  Be suspicious of areas where the readings are consistently higher - are there electrical/electronic accessories nearby? 

The point is the AM radio and the R2LC meter measures any energy at the selected frequency.  It does not tell you if it is useful energy.  Stated technically, you are measuring the sum of the signal (desired) plus noise (undesired).  By measuring around the layout with the TMCC base turned OFF you get a proxy for the noise level.   In terms of reliable data transmission, you want a high ratio of signal relative to noise....or what's commonly referred to as signal-to-noise (S/N) ratio.  So when you get a high reading at some location on the track you need to be sure this from desired (TMCC transmitter) signal level and not from the noise level; that's why it's important to first run a test of the noise level with the TMCC base turned OFF.  As you astutely point out, LC+ to the rescue!

cjack posted:

Like garage door openers then. Wonder how they decide where to put the frequency...FCC rules?

As a note, my alarm uses the 2.4GHz band.  

As to how the various frequency ranges are chosen, it tends to actually be a LACK of FCC rules.  The bands that are used for consumer devices are popular for these devices because manufacturers are free to use them with out needing a license.  You can thank your microwave oven for your Wifi and bluetooth and everything else on the 2.4GHz band.  The magnetron inside a microwave radiates noise at 2.4GHz, and because of this, the FCC determined it was a garbage band that was not of any use for  anything that needs to be regulated.  Amazing what can happen when you free just 500 megahertz of air space from regulation.  

Other bands have similar stories where the area was determined to be of no use to anyone, so left unregulated.  The old stand-by of 27MHz was popular for R/C control as well, with six dedicated rc channels mixed between the 40 CB radio channels that occupy the same area.  This was most likely left open for public use as it falls just below the world standard 10 meter amateur radio band.  I know for a fact every 27MHz device in the house is useless when My mother would fire up the Ham radio at 10 meters.  Also drove the neighbors nuts.  

You can see the entire regulated spectrum here:  https://transition.fcc.gov/oet...m/table/fcctable.pdf  It's not actually very useful for the average person, but fun to look through to see all the stuff going through the air around you.  

JGL

Gregg, there really is not a way to turn the alarm off, but given what everyone has said, the operating frequency is not in the range that will bother the layout.

George, I have been trying to redraw the track plan into its final configuration. The upper deck plan is sitting on my drafting table about 70% complete. The lower deck redraw hasn't been started, but the original plan is fairly accurate. There are no curves on the mainline under 072, and most are wider. There are no curves less than 054 anywhere. This really doesn't have anything to do with the signal issues, except that it would be nice to have a paper copy to map the test results.

Photo Jul 26, 10 58 31 PMIMG_5705

Stan, I'm going to gather all the components to make the meter car, then I can do a step by step assembly here with photos, in case anyone else wants to build one. I will probably need a little coaching to make the right connections. Thanks, that bit about LC+ just popped into my head. I do have my moments once in a while. Fortunately, they're not all "senior" (yet).

Attachments

Images (2)
  • Photo Jul 26, 10 58 31 PM
  • IMG_5705
stan2004 posted:
gunrunnerjohn posted:

Correct Dale, my point was simply that it was a DC measurement when you look at it.  Their sample application does have the resistor and cap.

Earlier you said:

I recall them saying when the signal was in the 40's, everything was great.  When the signal dropped down past about 30 and lower, the wheels started to fall off. 

What are the units?

The 455 kHz IC receiver meter output is a current proportional to signal strength with a specified nominal uA per dB slope.  The 51k resistor converts the units to Volts per dB.  The 0.1uF capacitor smooths the voltage.  For average signal strength you want to set the meter to DC Volts.   It's puzzling that the photo shows the meter set to AC Volts?

SS%20Car

 

Hi Guys

Been off from the forum for a week and just picked just up this thread on TMCC/Legacy signal and saw the picture of the earth ground signal test car which I built at the NJ-Hi Railers.  When taking earth ground signal readings you need to be on u amp setting.   When I originally took the picture, I didn't pick up the meter was set for ac volts.   This test car has been a "EXTREMELY USFULL" tool in addressing the TMCC/Legacy signal issues we had on our 30' X 200' layout, as it will indicate where you have week earth ground signal.  Also, when you attempt making changes to improve signal strength, you know right away if what you are doing has helped or not.  I would recommend anyone who is having serious earth signal issues (when you put your hand over the engine, your engine head light goes from nothing/blinking light to a solid on light)  to build one of these earth ground signal test cars which consists of a radio board, resistor & a capacitor.    I posted once before the forum complete plains of making one of these, and if anyone wants I can repost it.  

The other important item, is testing the other part of the TMCC/Legacy signal which comes from the "U" terminal of the command module.  Here we used as mentioned above, an oscilloscope to test signal strength and visually see if the signal was being corrupted by the layout.

Bob D

NJ-Hi Railers

 

Last edited by rad400

Bob, welcome to the conversation. That would be great if you would repost that here. I'm 99% sure I saw it and marked it, but it might take some digging, as I read and mark all topics related to TMCC signal here on the forum, and even some off. I was just gathering the materials for it now. The last piece missing is the car itself. I suspect I will be choosing a lighted caboose as my victim to donate its chassis to the cause.

At first I thought that the R2LC was going to come out of an early TMCC Dash 8, but when I opened it up, it was the old style electronics which predate the plug in boards. One thing caught my eye when I was in there, that was the antenna. New engines have their antennae just below the top of the shell in the form of a plate or a foil strip actually stuck to the plastic.

This unit had its wrapped around the inside of the nose of the cab, much lower than what we see today. The reason I bring this up is, of all my engines, I only have a few of these early units. I don't like the pulmor motors and the chunky bodies. However, one such unit seems to be immune to these signal issues. This can be seen in a couple of GoPro videos shot on my layout. Could the antenna placement be part of the secret to its success?

I'd be interested in the plans too.  While I have not experienced signal issues with Legacy ( knock on wood ) so far, that's not to say I'm in the clear yet.

Regarding the use of a portable radio for detection.  I hear the noise that Legacy produces over my stereo tuner in the layout room.  Is the detection procedure thus... a) shut down Legacy and scan the room with the portable, listening for any RFI set up by appliances, lighting, etc.?  and b) turn on Legacy and scan all areas with the portable, listening to the RFI set up by Legacy to see if it lessens in poor signal areas?

Bruce

Bob, I would think that you would either use the 51K resistor to turn the current into a voltage that is read with DC volts, OR

Use only the capacitor and read the current directly with the meter set to microamps,

But not both resistor and current measurement.

Perhaps the meter's low resistance in the current measurement mode just swamps out the effect of the 51K resistor (which is in parallel with the meter), hogging almost all the current.

In the chip's data sheet there is a designation near the RSSI meter that is 100KV, which might have been an attempt to say that the RSSI meter movement should be a 100K ohm/Volt meter - more properly written 100K/V.  A 100K/V meter that is 5V full scale would be a 50 uA movement.

None of this is critical for us if we just chose a circuit configuration and all use it.  I suggest the 51K, .1 cap and the meter set to 2V DC.  My gut feel is that we aren't going to get anything stronger than 2 volts for our whimpy pickup antenna. 

(An improvement might be to use a ferrite-rod "loopstick" connected between the antenna input and ground to get a better signal, but this would also introduce a lot of directionality to the signal pickup, which might be different than the antenna plate of a diesel or the handrails of a steamer.)

FWIW, I used a 9V battery to power my R2LC.  That way I just need to run a ground wire to the trucks - no rollers required.  I just realized that I also used one of the 2411 PW metal flatcars, which might have a different characteristic than a plastic gondola.  I did take care to rigidly mount the antenna, which is a chunk of brass rod.

I did a search and found the following post with what I believe are the aforementioned notes.

https://ogrforum.ogaugerr.com/t...48#48874849343058248

I took the liberty of copying the key diagram:

Signal CAR

I agree with Dale's comment about using the meter's Current measurement mode.   The Harbor Freight meter probably uses a 1k Ohm burden resistor for the 200uA range.  So as Dale says, this swamps out the 51k resistor and you are left with a 1k Ohm load and a voltmeter.  In any event this answers my earlier question of the scale when below 30 is bad, above 40 is good.  The meter could have been in the Voltage measurement mode without the 1k resistor swamping out the 51k resistor. 

The technical considerations get somewhat bogged down with the filtering effect of the 0.1uF capacitor when there's a 1k vs. 51k Ohm parallel resistor and how this plays off with the integrating algorithm of most hobby-grade voltmeters.

Bottom line is to simply make note of what measurement setup you're using so numbers from different set-ups can be normalized/compared.  I believe that if/when GRJ or JGL implement their wireless transponding signal-strength system, it will measure voltage (not current) as that's what the Arduino can do directly.

Attachments

Images (1)
  • Signal CAR

Bruce, basically just step a).  We are sniffing for noisy lights, switching power supplies, or leaking/arcing capacitors on the AC line inputs to most electronic devices.

I once had an AM noise problem (not with TMCC/Legacy, but just my AM radio) that was due to arcing contacts on a neighbor's aquarium heater.  The noise would cycle on and off as the heater turned on and off.  (We share the same power transformer.)

A somewhat related problem is desensitizing.  If there is a strong unwanted signal that gets into the front end of a radio receiver, that signal can activate the Automatic Gain Control circuitry that normally avoids overload with loud (desirable) signals.  In this case, the gain is also turned down for our wanted signal, desensitizing the receiver.  The R2LC has a tuned input stage that narrows the reception band to keep out unwanted signals, and there is no AGC.

Dale Manquen posted:

Bob, I would think that you would either use the 51K resistor to turn the current into a voltage that is read with DC volts, OR

Use only the capacitor and read the current directly with the meter set to microamps,

But not both resistor and current measurement.

Perhaps the meter's low resistance in the current measurement mode just swamps out the effect of the 51K resistor (which is in parallel with the meter), hogging almost all the current.

In the chip's data sheet there is a designation near the RSSI meter that is 100KV, which might have been an attempt to say that the RSSI meter movement should be a 100K ohm/Volt meter - more properly written 100K/V.  A 100K/V meter that is 5V full scale would be a 50 uA movement.

None of this is critical for us if we just chose a circuit configuration and all use it.  I suggest the 51K, .1 cap and the meter set to 2V DC.  My gut feel is that we aren't going to get anything stronger than 2 volts for our whimpy pickup antenna. 

(An improvement might be to use a ferrite-rod "loopstick" connected between the antenna input and ground to get a better signal, but this would also introduce a lot of directionality to the signal pickup, which might be different than the antenna plate of a diesel or the handrails of a steamer.)

FWIW, I used a 9V battery to power my R2LC.  That way I just need to run a ground wire to the trucks - no rollers required.  I just realized that I also used one of the 2411 PW metal flatcars, which might have a different characteristic than a plastic gondola.  I did take care to rigidly mount the antenna, which is a chunk of brass rod.

Dale

I am sure there are several ways this can be implemented.  I just followed what the specs indicated for the MC3372 IC, to get a RSSI output, using a 51K resistor & .1u cap.  It gave us at the club what we needed, which was a way to measure signal strength as we tried different ways to improve the earth ground portion of the signal and get real time results.

Also, this all originally came from a post that Gary Emmich posted over a year ago on how to get signal reading off of the Lionel radio board, which I took the info and made up the signal car.  

 I can't find my original write up, so putting the pieces together again as we speak and I will sent out a post shortly with all the information I have on the signal car.

Bob D

Bruce, as Stan pointed out a little further up the page, the AM radio and the signal detection car are doing a similar function. The advantage of the car is it can follow the track and quantify the signal. The AM radio is able to go anywhere, and sniff for other noise sources. It only registers presence, not level.

That's interesting about you hearing the Legacy noise on your stereo tuner. Out of curiosity, are the Legacy and the tuner plugged into the same house circuit? Could the noise be feeding back through the wall and coming out your speakers? When I was a kid, I remember the vacuum cleaner messing up the TV signal when it was on. I'm looking for a connection, but mostly shooting in the dark.

John, my electric bill is just fine. There would have been no way I could have lit the layout with anything less than LED bulbs. They use 9 Watts per 60 Watt equivalent bulb. The whole room takes less than 20 Amps to light, and generates almost no heat. They run, on average, maybe a couple hours a day. Nothing compared to heating water in the hot tub, which is why I turned that thing off, probably for good.

Add Reply

Post
OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653
330-757-3020

www.ogaugerr.com
×
×
×
×
Link copied to your clipboard.
×
×