Is my Zephyr working?

Can I get some thoughts on this situation. I have a HO layout (about 200 feet of track) and use a Zephyr. Just today I tried to run some locos and they just don’t get up to a normal speed. They run verrrrry slowly at maximum. Now I tried only 1 locomotive at a time to make sure I wasn’t overloading the works and they all did the same. The sound does work on all of the ones I tested. So I used my multimeter and found out I was getting only .04 volts (at maximum) on the 20 volt DCV scale from the rails.There was no variation as I adjusted the speed, but did go to 0 when I shut it off. Now I’m no techie so I don’t know if I’m even testing it correctly. I tried the quarter test and yes it did short out as it should. I then also checked the track and made sure there were no breaks or metal shorting the rails.

I can also run DC locos on my layout. So I disconnected the DCC and plugged in the DC Tech 3 power command and the DC locos ran fine. I’m thinking there is a problem with the Zephyr. I checked the output at the leads and get the same .04 V result.

Can anyone advise me on what the problem could be?

Thanks,

Tom

Select address 0 and set the speed to 0. Then dispatch this address. Now use your meter on the 20VAC range and measure the voltage from one of the track terminals to the ground terminal. Repeat with the other track terminal. Make sure the track power is turned on when you do this. The two numbers should be nearly identical. If one reads 6.5, the other should read 6.5. If one reads 6.5 and the other reads close to 0, then you probably have a blown output transistor in the Zephyr and it will need a vacation to Florida for repairs.

–Randy

Thanks for the reply.

Let me know if I am doing this correctly.

The lowest AC range my meter reads is 200. So I used this range and put the speed to zero. I connected the leads from my multimeter to the outputs(rail A and rail B) on the Zephyr and get 8.6 even when I reverse them.When I do this from the track I get slightly less , like 8.2. Am I doing it right?

Thanks,

Tom

You are doing it right… partly. Now measure from the A output to ground and the B output to ground.

Ok. I just did that. Both read 15.2. So does that mean the Zephyr is working as it should? If that’s the case what are my other considerations?

Thanks again.

Tom

The first guess would be yes. I am a little puzzled, maybe my brain isn’t working right, as I’ve been in bed with the crud for a couple days. I would have expected the ground to rail readings to be about half the rail to rail reading, rather than the other way round. Makes me suspect some kind of load on the layout. I’d be tempted to take the unit off the layout, and set up a simple test track, maybe just an oval to avoid crashes! I think I would pick one loco and work with it. See how it runs, and then reset the decoder to see what happens.

Sounds like it’s working ok, I second the idea to just hok the Zephyr up to a single piece of track or something simple and verify a loco runs that way.

–Randy

If each side measured 15.2 that’s way too high. Make sure that address 00 is set to 0 speed and dispatch it. Set your meter to DC voltage, place the positive probe on Track A and the negative on Ground. Record the reading. Do the same between Track B and Ground. Add the two readings together and that’s track voltage. For Zephyr with a PS315 power supply it should be about 13.5 - 14 volts.

The meter should be set on AC, not DC. On DC you should see something very close to zero. I’ve got a feeling there was something amiss on the 15.2 readings. I’d like to see everything repeated, very carefully, on an isolated track.

No, he’s measuring one side at a time, that’s done on the DC setting.

http://tsd.digitrax.com/index.php?a=1388

Martin Myers

Well, I see the words, and I realize it is a Digitrax page, but I am inclined to think it is a typo. Either rail should be spending half of its time positive relative to ground, and half the time negative. On DC, that’s going to be zero. I’m trying to think of what the ground reference could be to give a positive voltage on DC, and I can’t see it. The two rails should be going between plus and minus sixish, relative to ground (assuming ground is halfway between the rails. Relative to each other the reading should be about 12. If I’m wrong, I’ll eat my hat, or at least another Airborne!

Looks like we’re both right.

I’ve been using the DC setting for years. The total always matches my RRampmeter readings of 13.9 volts on my Zephyr.

Just for the heck of it. I checked it on the AC setting and got the same total. I also got 13.9 volts between Rail A and Rail B. It’s a true RMS meter though.

Martin Myers

I think a DC reading from a garden variety meter would be near to zero. The RRAmpmeter is definitely the right tool for the job. I can’t get my head around the DC setting getting a reading, though if it was a clever enough meter, I guess it might!

I hope your hungry! Actually, some systems may operate the way you describe, but Digitrax doesn’t. With Digitrax, assuming 12 volts rail to rail in our example, each rail relative to ground alternates between zero and 12 volts, when rail A is at 12, rail B is at zero, and when rail A is at zero rail B is at 12 volts. This gives the same output on the rails relative to each other as it would if each rail was alternating between + and -6 volts, but since each rail is always positive relative to ground, you can use a multimeter set to DC to measure the voltage on each rail to ground.

Measuring DC voltage from rail to ground is only an approximation. AC voltage is off as well as meters are designed for 60 cycle sine waves and not the 8KHz square waves of DCC.
Also you have to consider whether you want to measure peak voltage or average voltage (average more useful in diagnosing problems.

Some suggestions on very simple rectifier circuits and procedures can be found at:
’s DCC Meters page
OR
Allan Gartner’s Wiring for DCC

John Balogh has provided a general purpose DCC adapter based on a NCE design.

OK, I can see that this morning! Ground is just an arbitrary point, for some reason last night it wasn’t making sense. So you could measure it as AC or DC and get some kind of meaningful reading. So, in this case, I think it still makes sense to put the system on a simple oval, or even a single track, and see what’s up.

I connected the leads to a separate piece of flextrack and the results were similar. I guess I’ll have to give Digitrax a call and see what they say.

Thanks for all of your help.

Tom

OOPS!

How embarrassing. I’ll bet this has never happened to anyone before. LOL. I did a little investigating of my power supply. Well I have a power bar under by bench onto which I have a few items connected. One is my Digitrax power supply and the other is a simple Black & Decker power tool recharger 12 Volts 210 ma…Guess which one was connected to my Zephyr? Ya…the B & D. I was surprised it actually fit into the back of my Zephyr, but it did and I just didn’t notice as they are almost identical. Of course after I connected the correct one everything works fine. I’m fortunate I didn’t fry the Zephyr.

So thanks everyone for trying to help me. Sorry for having you guys go to so much research for me when it was my fault.

Tom

Actually, if you measure the DC voltage of each rail to ground and add them together it WILL give the correct rail to rail voltage, or at least very close to it. Most meters simply measure the average voltage. Since each output spends half it’s time at full voltage and half it’s time at zero volts(assuming we are not using zero stretching), then the average voltage on each output will be half of the rail to rail voltage. The reason most meters do not read the correct voltage when you measure from rail to rail on the AC setting is because the meter is expecting a sine wave. Again, all the meter is actually measuring is the average voltage, but it is adjusting this voltage(basically multiplying it by about 1.11) to show what it believes to be the RMS voltage.

Actually this led to a good discussion on measure voltages in DCC, so not a waste of mine time (at least) and to what appears to be helpful links[:D].
Alan