Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Using the PR3 to program Sound Loco's

10264 views
31 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Friday, August 20, 2010 3:04 PM

Well I finally received my replacement unit today.  Fired it up and it worked like a charm right away.  Shame I had to spend that time playing with the defective unit but all's well now. Thumbs Up

Springfield PA

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Saturday, August 7, 2010 4:17 PM

well it's back and none in stock.  I'll have to wait a week or so for the replacement. Sad

Springfield PA

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Friday, August 6, 2010 9:24 PM

That's what it looks like. The basic track drive is working but the signal doesn't look like it's being injected.

I'll take my laptop with me tomorrow to lins and hopefully they'll have another one in stock. I'll test it before leaving since it's a 30 mile drive.

Springfield PA

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Friday, August 6, 2010 9:11 PM

 Kinda suprised you get a reading on the program track because the voltage is only applied in quick pulses. However what readings you do get should be similar regardless of which device is programming. SOunds like a dead PR3, on the track drive for the standalone mode.

                            --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Friday, August 6, 2010 8:17 PM

rrinker

 This test is not a red herring - you just proved that the communications between your computer and the PR3 are fine. Regardless of which outputs the PR3 uses, the computer side circuitry is the same regardless if it's used in standalone mode or MS100 mode. So, the drivers are properly installed, the cable is good, the port settings are valid.

 Now, connect the power supply and connect the track, and use the setting in JMRI to change the mode - DON'T use the PR3 buttons - JMRI can switch back and forth by itself. If all you really have are the QSI locos, try opening the proper docoder manually in JMRI and programmign somethign obvious - like maybe turn the horn volume to the lowest setting or something, just to see if it can write the CV. It may write without reading - in which case it could simply be the OEM QSI's inrush issue. The new aftermarket Revolution decoders I had no problem with on a plain old program track. It's still like to see a test with a common brand non-sound decoder, one that is not known to have issues like the Bachmann, before declaring the program track output of the PR3 is dead and it needs to be sent back. You could try the 1K resistor trick with the Bachmann, put a 1K resistor across the rails and see if you can program the decoder but I still wouldn;t consider that a true validation until a decoder that is known to work with no funky issues either works or does not work. If a know working non-sound decoder also fails to work on the program track, then I would declare the PR3 defective. The fact that it can tell the difference between no loco on the track and a decoder read error tends to tell me the program track works fine, although I suppose it IS possible that it's only slightly damaged and can 'see' somethign from the decoder but not enough and can still tell the difference between an open circuit and the fact that something is there.

 REALLY crazy - if using a laptop, try it with the power supply disconnected (battery power) or if possible try reversing the AC plug on the power supply (if it's not polarized). I'm thinking possible ground loop issue since the PR3 is not optically isolated. It wouldn;t be the first time this has come up! (someone on the JMRI list had a similar problem and tried for weeks to get a PR3 to work on his Netbook under Windows - it worked on his Mac just fine)

 

                                                     --Randy

 

Thanks

I pretty much covered those items including the power plug on the laptop.

I went ahead and hooked up the program track to the DCS100 and kept the PR3 in the DCS100 mode and tried a few loco's Both Atlas loco's and a paragon 2 read and write perfectly.  I measured the voltage to the track with an analog meter in the AC scale and read 6 volts.

I then switched the wires back to the pr3 and switched to stand alone mode. The voltage does not read in the AC scale but does read in the DC scale when doing a read operation.  Something there is definitely different. Almost like the DCC is driving the track but the data is not switching on.

I'll play with it some more but it looks like something is up with the PR3's program track ckt.  If there are any more idea's I'll try them as well. 

Looks like a trip to Lins Junction tomorrow otherwise.

Springfield PA

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Friday, August 6, 2010 6:40 PM

 This test is not a red herring - you just proved that the communications between your computer and the PR3 are fine. Regardless of which outputs the PR3 uses, the computer side circuitry is the same regardless if it's used in standalone mode or MS100 mode. So, the drivers are properly installed, the cable is good, the port settings are valid.

 Now, connect the power supply and connect the track, and use the setting in JMRI to change the mode - DON'T use the PR3 buttons - JMRI can switch back and forth by itself. If all you really have are the QSI locos, try opening the proper docoder manually in JMRI and programmign somethign obvious - like maybe turn the horn volume to the lowest setting or something, just to see if it can write the CV. It may write without reading - in which case it could simply be the OEM QSI's inrush issue. The new aftermarket Revolution decoders I had no problem with on a plain old program track. It's still like to see a test with a common brand non-sound decoder, one that is not known to have issues like the Bachmann, before declaring the program track output of the PR3 is dead and it needs to be sent back. You could try the 1K resistor trick with the Bachmann, put a 1K resistor across the rails and see if you can program the decoder but I still wouldn;t consider that a true validation until a decoder that is known to work with no funky issues either works or does not work. If a know working non-sound decoder also fails to work on the program track, then I would declare the PR3 defective. The fact that it can tell the difference between no loco on the track and a decoder read error tends to tell me the program track works fine, although I suppose it IS possible that it's only slightly damaged and can 'see' somethign from the decoder but not enough and can still tell the difference between an open circuit and the fact that something is there.

 REALLY crazy - if using a laptop, try it with the power supply disconnected (battery power) or if possible try reversing the AC plug on the power supply (if it's not polarized). I'm thinking possible ground loop issue since the PR3 is not optically isolated. It wouldn;t be the first time this has come up! (someone on the JMRI list had a similar problem and tried for weeks to get a PR3 to work on his Netbook under Windows - it worked on his Mac just fine)

 

                                                     --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Friday, August 6, 2010 5:02 PM

Just hooked it up to the layout and ran monitor mode and it fired right up.  Activity is showing fine.  I also opened up a throttle and it ran a loco fine. 

Springfield PA

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Friday, August 6, 2010 4:01 PM

Stevert

Hamltnblue

Thanks for the info

I'm starting to think I might have a bad unit.

I loaded the drivers, the sound downloader, and decoder pro on a windows XP laptop and am getting the same error code.  I also tried matching the baud rate at the decoder pr 57600 and with flow control on and off with no luck.  I also tried a dc power supply that's putting out 17.5 volts under load.

  I'm not sure what the problem is, but I don't think it's the PR3.  You said earlier:

 

Hamltnblue
I also noticed that If there is no loco on the track and I try to read a CV I get a 301 code saying no loco. If  a loco is on the track I get a 308 code

  If the PR3 can tell the difference between no decoder (error 301) and no ack from the decoder (error code 308) and feed that info back to JMRI, the computer is obviously talking to the PR3, and the PR3 can obviously tell that something is or isn't on the programming track.

  Are you sure you've covered all the basics: All the one's Suggested and the usual suspects, including changing the USB cable, Higher voltage power supply, Port settings, Running in windows XP mode,  Trying different computer with windows XP, Kicking the dog, etc.

- Program track completely isolated from everything except the PR3?

Yes the program track is a length of flex sitting in front of my computer monitor. The track wires are soldered on.  Voltage is 0 and raises when the unit is trying to read or program the loco.  My paragon 2 will even run it's sound when the track voltage comes on. Neither program or read appears to work.

- Good, tight connections between the PR3 and the programming track?

Yup.  maximum allowable torquage.

- Clean programming track/loco wheels?

The used brightboy and 91% alcohol are right in front of me (after being used)

 

- Voice acknowledgement turned off on the decoders that support it?

Not on the Atlas butthe other 3 loco's had no luck either.

Testing it in MS100 mode is a red herring here, because it uses different output circuitry for the LocoNet than it does for the programming track outputs.

That's next.

Springfield PA

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Friday, August 6, 2010 3:55 PM

rrinker

 QUick sanity check, put it in "MS100" mode and connect it to your command station. In JMRI make sure it's also set to MS100 mode and not standalone. Open the Locnet monitor and run a train with your throttle (doesn't even have to be a train on the track, just so the system is up and in normal run mode), and if it's working OK you should see the Loconet data in the window as you change speed or hit function keys. If you don;t get that, it's not communicating. Also try the track power on/off utility in JMRI and see if the track power goes on and off.

 Thought I could quickly check with this laptop since it's the one I used with the PR3, but naturally since there is no device plugged in, nothign shows in Device Manager. But a common issue on older versions of JMRI was tha tthe Java com libraries never actually changed the windows settings. The fix was to go into device manager and look at the properties for whichever com port the USB device got assigned to, and make sure in THERE is it set right. 57600 8 N 1 should be right, or maybe 115k. THEN go into JMRI and make sure it is using the same values.

                                            --Randy

 

I'll try to do that next.  As far as the prot settings I tried it in default as well as 57600 8N 1 with hardware control on and off (set on both sides) No luck.  The JMRI is the latest version.

I've also tried slowing the USB port to USB 1.0 mode with no luck.

 

Springfield PA

  • Member since
    December 2001
  • 1,932 posts
Posted by Stevert on Friday, August 6, 2010 3:47 PM

Hamltnblue

Thanks for the info

I'm starting to think I might have a bad unit.

I loaded the drivers, the sound downloader, and decoder pro on a windows XP laptop and am getting the same error code.  I also tried matching the baud rate at the decoder pr 57600 and with flow control on and off with no luck.  I also tried a dc power supply that's putting out 17.5 volts under load.

  I'm not sure what the problem is, but I don't think it's the PR3.  You said earlier:

 

Hamltnblue
I also noticed that If there is no loco on the track and I try to read a CV I get a 301 code saying no loco. If  a loco is on the track I get a 308 code

  If the PR3 can tell the difference between no decoder (error 301) and no ack from the decoder (error code 308) and feed that info back to JMRI, the computer is obviously talking to the PR3, and the PR3 can obviously tell that something is or isn't on the programming track.

  Are you sure you've covered all the basics:

- Program track completely isolated from everything except the PR3?

- Good, tight connections between the PR3 and the programming track?

- Clean programming track/loco wheels?

- Voice acknowledgement turned off on the decoders that support it?

Testing it in MS100 mode is a red herring here, because it uses different output circuitry for the LocoNet than it does for the programming track outputs.

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Friday, August 6, 2010 3:26 PM

 QUick sanity check, put it in "MS100" mode and connect it to your command station. In JMRI make sure it's also set to MS100 mode and not standalone. Open the Locnet monitor and run a train with your throttle (doesn't even have to be a train on the track, just so the system is up and in normal run mode), and if it's working OK you should see the Loconet data in the window as you change speed or hit function keys. If you don;t get that, it's not communicating. Also try the track power on/off utility in JMRI and see if the track power goes on and off.

 Thought I could quickly check with this laptop since it's the one I used with the PR3, but naturally since there is no device plugged in, nothign shows in Device Manager. But a common issue on older versions of JMRI was tha tthe Java com libraries never actually changed the windows settings. The fix was to go into device manager and look at the properties for whichever com port the USB device got assigned to, and make sure in THERE is it set right. 57600 8 N 1 should be right, or maybe 115k. THEN go into JMRI and make sure it is using the same values.

                                            --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Friday, August 6, 2010 2:16 PM

Thanks for the info

I'm starting to think I might have a bad unit.

I loaded the drivers, the sound downloader, and decoder pro on a windows XP laptop and am getting the same error code.  I also tried matching the baud rate at the decoder pr 57600 and with flow control on and off with no luck.  I also tried a dc power supply that's putting out 17.5 volts under load.

 

Springfield PA

  • Member since
    December 2001
  • From: Trois-Rivieres Quebec Canada
  • 1,063 posts
Posted by jalajoie on Friday, August 6, 2010 10:51 AM

When you will install the PR3 on Win XP, make sure the com port you use is the one windows indicate in the peripherique device manager. But you may have already done so.

Jack W.

  • Member since
    December 2004
  • From: Bedford, MA, USA
  • 21,483 posts
Posted by MisterBeasley on Friday, August 6, 2010 10:01 AM

Have you tried the PR3 with the SoundLoader program from Digitrax?

I have a PR2.  It works with SoundLoader, but I've never been able to make it work with JMRI.  I am using an old MRC power supply, and people have suggested that it's not strong enough for the PR2.

Anyway, the simpler SoundLoader program may help you make sure your COM ports are set up correctly.

It takes an iron man to play with a toy iron horse. 

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Friday, August 6, 2010 5:36 AM

Forgot to add, I also swapped the USB cable with my printer and no change.

 

Springfield PA

  • Member since
    June 2006
  • From: Maryville IL
  • 9,577 posts
Posted by cudaken on Friday, August 6, 2010 5:15 AM

  I would all so pick up a new USB cord as well.

I hate Rust

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Thursday, August 5, 2010 9:34 PM

Thanks

I have only 2 non sound decoders and they;re both bachmann.

About half are Atlas Golds and the rest are Tsunami's Broadway limited, and MTH.

Tomorrow I'll bring my laptop from work home since it has windows XP installed and give it a try.

It seems weird that it's going throught the paces and nothing is programming or reading.

I guess I'll have to try to find a higher voltage power supply as well.

 

Springfield PA

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Thursday, August 5, 2010 8:58 PM

 Got anything with a plain NCE, Digitrax, or TCS decoder in it?  The Bachmann dcc on-board decoders are notoriously troublesome.  I haven't tried much since I just set up my program/test track, but I had no problems with one of my P2K GP-7's with a TCS decoder on the PR3. My Zephyr has programmed everything I've tried so far, but I've never tried a Tsunami. It did handle QSI and Loksound, and the old Sountraxx LC decoder I had, as well as Digitrax, NCE, and TCS non-sound decoders.

                                                 --Randy


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Thursday, August 5, 2010 3:23 PM

I've since tried a different power source which is 13.5 volts DC with no change.

I also noticed that If there is no loco on the track and I try to read a CV I get a 301 code saying no loco. If  a loco is on the track I get a 308 code

 

Springfield PA

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Thursday, August 5, 2010 2:34 PM

After playing some more I'm getting error code 308 which is no ack from loco.

I tried a bachmann 70 tonner without sound and getting same thing

I also tried one of my soundtraxx loco's and when trying to identify I get the 308 but the PR 3 has a high pitched noise while trying to read.

As it sits now the decoder pro is set for PR3 stand alone mode. The pr3 USB light is one second on and one second off. The stat light flashes about every 2 seconds. When trying to read a decoder it flasher faster.

The com port settings are set to the current port which is 4, matched th57600 baud with hardware flow control on to match decoder pro.  I don't see anywhere that it mentions the correct data bit, parity and stop bit.  The port is set for default N,8,1

 

Springfield PA

  • Member since
    February 2008
  • 8,878 posts
Posted by maxman on Thursday, August 5, 2010 2:11 PM

Hamltnblue

Maybe I was using the wrong version of QSI decoder.  I was trying version 7 when maybe it should have been 6.  I'll have to try that tonight.

The loco is the MP15DC.

When I looked up the MP15DC in DecoderPro to answer someone else's question, I found it listed under QSI version 6.

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Thursday, August 5, 2010 1:46 PM

Thanks all 

I Just got home and will try what has been suggested today, starting with cleaning the track.

 

Springfield PA

  • Member since
    December 2001
  • From: Trois-Rivieres Quebec Canada
  • 1,063 posts
Posted by jalajoie on Thursday, August 5, 2010 9:34 AM

What is the error number you are receiving?

If it is a reading time out error, 99% of the time it is from a wrong com port in the set up.

Other errors may be due to a faulty USB cable or port, I have seen this more than once.

 

Jack W.

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Thursday, August 5, 2010 9:25 AM

Thanks

The version is the latest one which is 2.10

I tried the Ident Decodeer and was receiving errors.

Th epower supply is the Digitrax PS14

I haven't tried a non sound decoder yet.

 

Springfield PA

  • Member since
    December 2001
  • 1,932 posts
Posted by Stevert on Thursday, August 5, 2010 8:54 AM

Hamltnblue

The software I am running (And learning) is decoder pro.  I have it set to the PR3 in stand alone program mode. It looked like it was downloading ok to the decoder but I didn't have time to test it.  It just wasn't reading it.  Maybe I was using the wrong version of QSI decoder.  I was trying version 7 when maybe it should have been 6.  I'll have to try that tonight.

The loco is the MP15DC.

Which version of JMRI?  Some changes to the QSI definitions were put in about 2.9.5 to fix an issue with reading QSI V7 decoders.  I'd really suggest you upgrade to 2.10 if you haven't already.

And if you use the "Ident decoder" button in JMRI, it will read the decoder information and present you with the correct decoder definition.  I've found that it's very accurate in correctly identifying the particular QSI version you're reading.

Also, you never did tell us what you're using to power your PR3.  The originally-recommended PS12 was pretty anemic, and the newer PS14 is better.  But for sound decoders, something in the 16v to 18v range is better still (the PR3 can handle up to 20v).  I use a 1.1 amp, 18V power brick from an old printer with my PR3. 

  • Member since
    June 2006
  • From: Maryville IL
  • 9,577 posts
Posted by cudaken on Thursday, August 5, 2010 6:42 AM

  Can you read none sound decoders?

  Like I posted before, when I first hooked mine up all I could read where none sound decoders. Hopefully  Simon1966 will chime in due to the fact he fixed mine. It been a year ago, so I don't remember the details clearly. First he gave me a different USB cable. Then he went into Comfig and  assigned the PR3 a USB port that was on the front of the computer.

 Sometimes, if the wheel or test track is little dirty, I will get a error 308. 

 After you get it working, I am sure you will like it.

         Ken

I hate Rust

  • Member since
    December 2004
  • From: Bedford, MA, USA
  • 21,483 posts
Posted by MisterBeasley on Thursday, August 5, 2010 6:24 AM

The PR3 is a Digitrax product.  It can program other decoders, but only the standard functions.

It will download sound files into Digitrax SFX sound decoders.  It will NOT download sound files into any other kind of sound decoders, though.  So, you should be able to read and write CVs in your QSI decoder, but you can't change the sound files.

I may not be up to date on their latest products, but I don't think QSI has a user-downloadable sound decoder.

It takes an iron man to play with a toy iron horse. 

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Thursday, August 5, 2010 5:21 AM

The software I am running (And learning) is decoder pro.  I have it set to the PR3 in stand alone program mode. It looked like it was downloading ok to the decoder but I didn't have time to test it.  It just wasn't reading it.  Maybe I was using the wrong version of QSI decoder.  I was trying version 7 when maybe it should have been 6.  I'll have to try that tonight.

The loco is the MP15DC.

Springfield PA

  • Member since
    January 2001
  • From: Kansas
  • 808 posts
Posted by jamnest on Wednesday, August 4, 2010 9:56 PM

YES  I have an Atlas MP15 locomotive with the QSI sound decoder.  I have both a PR3 and MS100 with a Windows XP personal computer and JMRI Decoder Pro for programing.  No problems.

I also use the PR3 as a stand alone programer with my laptop, Windows Vista and JMRI Decoder Pro.  No Problems.

Jim, Modeling the Kansas City Southern Lines in HO scale.

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!