Hi all -- I am the "maintenance supervisor" for a fellow club members' trains, and I have an A-B-A set of Proto 2000 E7s with QSI sound/DCC that are giving me fits. (Yes, I know, that's what they are famous for.)
After sitting a while, they went dead. A web search suggested I reset the units. I did. All came back to life, but only the B unit had sound. It went on the railroad and I put the As aside.
Today I put one of the A units on a test track with an NCE controller and it was totally dead. Tried it on DC; totally dead. (We had originally programmed the units for no DC conversion.) Put it on another test track with a Zephyr -- and the sound system rumbled to life! I selected address #3, the lights came on and the sound stopped. It moved, but there was no sound.
Back on the NCE track and the engine responded to address #3, but was still silent. I tried resetting the unit (3 CVs), again it responded to address 3, but no sound. I was able to program it to its proper number with the NCE controller (4009) and make the long address active... but still no sound.
Clearly the sound function is still there, but I have no clue why it keeps going away. Any suggestions?
The QSI chip says the revision is 2009 -- 1-2009, I think.
I did not have the other A unit with me so I was unable to go through the same process.
And yes, I tried un-muting with F8... no dice.
Anyone ever dealt with anything like this before?
Thanks in advance!
Aaron
Autonerd I tried resetting the unit (3 CVs), again it responded to address 3, but no sound.
Hi,
I'm wondering if you tried the "Hard Reset" with the magnet? The reed switch is buried pretty deep in the E7 body and the supplied magnet may not be strong enough. You may have to remove the shell. The reed switch is right over the chip.
QSI_E7 by Edmund, on Flickr
You have to cut off the power then restore the power to the rails for the reset to take effect. I have heard of some with a bad reed switch so you would have to use a pair of grabber clips on a jumper wire to bridge the switch.
I've heard of at least one case where the reed switch was welded closed so the decoder was always in reset mode. If you changed the address, though, that seems unlikely in your case.
Don't forget that silly F6 sleep, and deep sleep mode, too.
http://cs.trains.com/mrr/f/744/p/242811/2703567.aspx
I know how frustrating it is. I have made successful QSI resets only by trial and error. Mostly error.
Good Luck, Ed
Ya know what would sell really well for maybe $20-30? A compleat book on fixing what goes wrong with DCC.
For example, my three T55 diesels just WOULDN'T go for it. I asked several times on forums, and got answers that didn't work. Then I happened to buy a program track booster. Problem instantly solved. "Mostly error."
Certainly a large chapter could be devoted to "moves but no sound".
And another chapter to "sound but won't move".
Two of MY favorites!
Ed
The QSI decoders can be a bit quirky in that some of them 'sulk' if they have received track power, but nobody called them over to play. I still have perfectly functional QSI from the first generation that will do that. I simply have to acquire the correct and current decoder address, and they'll come back to life.
Second point...QSI decoders have a three-stage induction to coma. It involves successive double-pressed of F9. When the power next comes to the rails, they remain dead, silent, or may let off a faint brake-pipe open cock sound, like air release. If you wish to have them active again, you simply double-click on F6.
Third point, already stated by the first responder above, is that for a full reset to take place, you must remove power to the rails after entering the third set of reset CV numbers. Then, restore power to the tracks, acquire Add "03", and the QSI should respond, but now at full volume and no CV2, CV3, or CV4 programming; they'll lurch into motion as soon as they get enough voltage, and stop on a dime, as examples. Also, the individual sounds will now all be at factory default levels. Including that horrible and annoying 'injector' wail which is almost impossible to hear trackside anyway.
If there is a reed switch, you can use that, or perhaps the booklets mentions a magnetic want do wave over the reed switch. That's only if the three-step CV reset fails, and it has never failed for me.
Seems odd that the Digitrax sound decoders in the T55 locos would need a program track booster.
But then, people sya they need a booster for lots of differnet locos that seem to work fine for me. Dunno what I do differently - other than not use JMRI to try and read 100+ CVs all in one go. I do keep the wires tot he program track SHORT, and relatively large - #18.
My Zephyr alone has issues changing the address on QSI, but when used with a DT40x throttle, it works fine, including the automatic setting of CV29. But there's a difference in the way it works - Zephyr alone just makes one attempt and that's it. On the DT40x, it asks if you want to enable the long address if setting a long address - and keeps repeating that until you stop pressing yes or no and let it time out, so after a few yes answers then I let it time out, and it has always worked.
I was going to say that it would be interesting to see what my one and only QSI loco, an Atlas Trainmaster, does after more than 5 years stored away - but I DID just have it out recently - fired right up, sounds all worked, and it still had the cab number as the functioning address. Again - what did I do that others do or don't do with their locos? I had it in the transport box I use to carry my equipment to train shows - the ones by A-Line, so it was sitting on its wheels. MORE than 5 years, actually, it's been that long since I moved and it was a year before I moved that I was at the last club setup. I will say I pretty much never use the F6 shutdown stuff - I suspect that is the OP's issue, it is simply in one of the shutdown modes - ALL the way down, it won't move, even though the address is correct. Another one, it moves but acts like a silent loco.
Only issue with my loco was it was a little jerky at first, but as soon as I ran it back and forth a few times it ran smoothly once again.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.