Skip to main content

Closed15 replies

Looking for some technical insight on how the chips on a PS1 soundboard work?

G GGGOGR Forum Member
I am interested in understanding the details of how the PS-1 sound board microprocessor work in conjunction with the chip in the PLCC.

Really just for my education, but partially to accurately describe problems related to "scrambled boards/chips". This may be a basic Computer Science type answer.

I have all the basics on the various issues, but after reviewing the documents and articles on the web about these issues I do believe scramble, memory, software and chip are being used very loosely and maybe not very precisely.

My questions are:

Where is the software stored? In the processor and the PLCC Chip or just one?

What type of chip is in the PLCC? Is this just a Read Only memory, or does it have data storage capability, is it a processor too?

Same question about the microprocessor? What type and how does it work in laymens terms?

How do they interact?

My questions are related to the 2 cases listed below. In both cases the sound board is fine but "software gets scramble". In one case you install another chip in the PLCC to resolve the issue and put the original chip back in. So this sounds like the software scrambled must be on the Soundboard somewhere, since you removed the PLCC Chip.

In the other case the "listed" emgines have a software issue that deslects the engine. The only way to fix this software is replace the PLCC chip. This would imply the microprocessor or software on the board is fine, but the chip got scrambled. How does the software glitch effect the PLCC chip. I know it has to do with low battery state, but if the software is on the board, how does the chip software mess up?

I actually have a listed engine in for repair. It was using the BCR. The board and BCR are fine, but the chip in the PLCC doesn't work. I can install other chips and the engine works fine. The solution is replace the chip, but I am trying to understand how the 2 cases differ, yet both are classified as "scrambled boards or chips" by various documents. Thanks, G

This is from the MTH web site.

"Owners of some M.T.H. Electric Trains' Proto-Sound 1 equipped engines built prior to mid-2000 may experience one of two software related errors that are initiated from low or discharged batteries. Some confusion in the marketplace regarding these errors has led to the replacement of the Proto-Sound board when correction could have been resolved through a software upgrade. In each case, the software error can be avoided entirely if the owner replaces the battery prior to use. Battery replacement should only be necessary if the locomotive has not been run within a year and the battery is no older than a year.

What follows is a description of each error and the necessary steps that must be taken to correct the problem. In case #2, the problem only occurs in engines produced in 1995 and 1996 as illustrated in the attached list. QSI, the developer of M.T.H.'s original Proto-Sound system corrected that problem in a later software version.

Case #1: Proto-Sound 1 locomotive starts up and reports three clank sounds

This scenario is always precipitated by a low or discharged battery and can usually be avoided by charging or replacing the battery prior to powering up and operating the locomotive. The problem doesn't occur as soon as the engine is powered up, but rather when the operator tries to get the engine to move by using the direction button or turning the throttle off and then on again to enter the forward or reverse direction. If the battery is too low or discharged at this point it will be unable to supply the circuit board with enough current to operate normally causing the board to become "scrambled". This can be confirmed with the sounding of the three clanks. At this point it will be necessary to reprogram the board with a special software chip. Once reprogrammed, the original software chip can be reinstalled and the engine should perform normally. Essentially, the special software chip reinitializes the board. Unfortunately, the problem can reoccur in the future should the battery become dangerously low or completely discharged. The problem stems from the QSI design of the Proto-Sound board and no permanent fix is available.

Case #2 Proto-Sound 1 locomotive starts up but will not move.

This scenario only occurs in some engines produced in 1996, 1997 and early 1998. The error occurs as a result of a low or discharged battery and causes the software to "deselect" the locomotive. Users of QSI's more full-featured sound systems may recognize the "selection" feature as the QSI system's ability to give engines an ID number. When given an ID number, each engine can be turned on or off (deselected) so as to respond or not to transformer-related actions. An engine that has been deselected will not turn on even if track power is on. The engine will remain quiet and still during all transformer actions until "reselected" through special whistle and bell commands from the transformer. The feature was a precursor to today's command control operations found in the DCS and TMCC systems. However, QSI's approach was deemed to complicated a procedure by M.T.H. for Proto-Sound 1 locomotives. As a result, M.T.H. asked QSI to eliminate the feature from the software used in Proto-Sound 1 engines. Unfortunately, a bug in the QSI software deselects the locomotive when the battery charge becomes too low resulting in an engine that starts up but fails to leave the Reset position. Users may mistakenly assume that the locomotive is locked in neutral. Replacing the battery and performing the unlock procedure or resetting the engine through Feature 18 will not resolve the problem.

Fortunately, the solution for #2 resides in a software update through the installation of a new chip that permanently fixes the "deselect" function from occurring in the future should the battery charge become too low or discharged.
Original Post

Replies sorted oldest to newest

G GGGOGR Forum Member
Chuck, Actually I did, including e-mails that someone sent me. Unfortunately it is vague enough to not really answer the question. The Scale Tin Rail does have a ton of info and I may have missed it, but I don't think they really address how the fault occurs with sufficent details or how the board components interact (processor and PLCC Chip). G
chuckchuckOGR Forum Member
Concentrate your search on early QSI documentation that dealt specifically with upgrades paths offered for earlier models of the QSI product line. There is some discussion of "genealogy" of PS-1 and how the product fit in. I think PS-1 would be close to QSI-1.25 (which never actually existed). PS-1 design was a compromise to accommodate MTH's wishes for more sounds and a desire (at that time) to drop the command/control aspects of the product for a variety of reasons.
I ironlake2OGR Forum Member
quote:
Originally posted by CRH:
Why don't you just get on the phone and call QSI and speak with either Pat Quinn or Fred Severson as they invented the stuff..well along with help from Neil Young? Smile


good luck to try and get them, they do not sell direct anymore and are entirely into making dcc sound sets etc for the the dcc market. I used to sell and repair their products when they were available for aftermarket sale and I do not know their phone numbers anymore.
the fiasco they went through with mth stopped the 3 rail sound sets all together. try and get product from art boyington. He has had a soundset on order for me for two years and no results.
G GGGOGR Forum Member
quote:
Originally posted by CRH:
Why don't you just get on the phone and call QSI and speak with either Pat Quinn or Fred Severson as they invented the stuff..well along with help from Neil Young? Smile


Actually was going to try that wise guy Wink I did find the QS Industries web site. There is a phone but no e-mail. There is another site also, but again low on contact info.

I was figuring an e-mail, were a response could be given on the CEO's time line, vice my call time would be a more thoughtful approach.

I also was hoping some forum expertise on the EPROM and Microcontroller could give me some help. Big Grin G
G GGGOGR Forum Member
John, Thanks. The QSI solutions seems to be all N/HO DCC stuff, and G.

I guess STR is the only remaining options. QS Industries states they are their tech rep. I have read a series of STR e-mails wrt to the history and techical aspect. Unfortunately a little long winded, and tended to promote upgrades an mods.

From the QS2+ manual it seems that if this goes to deselect, you need to address the specific Temp ID or Road Id and Eng ID to reselect it. Or do a select all.

The problem is that under this condition the engine can only get from reset state zero to reset state one. No further. I tried many combinations of horn press to hit an ID number, but either the glitch set an extremely high ID number greater than a 130 (I tried), or the software can't execute the select all function since I don't think MTH wanted those features at all. So the glitch deselects the engine, but no way to recover.

From the manual, clearly IDs can be saved, and they are retained in memory even when power is removed.

I will try the e-mail to STR and post answer if I get one. G
G GGGOGR Forum Member
While I don't have specific information from the inventor here is some additional data I have found from actual testing.

The PS-1 microprocessor and PS-1 Chip must both store temporary data.

I am working on several MTH PS-1 engines that are on MTH's list for being "deselected".

The engines will turn on and give 2 bells but no sound, no motion. Trying to cycle through the reset states doesn't work either. They will enter the first reset state (one clank), but will not advance beyond that and remain silent.

The engine works fine with a different test chip. Every combination of up and down of the factory reset chips will not get the "deselected" chip/engine to work.

Then I had a Weaver engine come in with the QSI QS-2+ system. This system does have the ability to select and deselect engines. This engine just had a dead battery. Otherwise it works fine.

So I took the deselected chip from the MTH Engine and placed it in the QSI QS 2+ system. The engine/chip combo worked fine!!

I ran the MTH chip through all the reset states and cleared any possible Temp IDs, Eng and Road IDs, then did a Reset #18.

The QSI QS-2+ chipped also worked in the MTH engine that was deselected.

So I put the original MTH chip back in the MTH engine and IT STILL DID NOT work.

I put it back in the QSI engine and it STILL worked.

So the chip seems to be good, but it will not work with it's original processor.

I then tried the QSI chip back in the original MTH engine and though it works, it too won't cycle out of the reset state.

So the only thing I can think of is the processor has memory bit activated that is preventing it from executing the original chips software or vice versa. A conflict of some nature.

My last test will be to try to use the original MTH chip in a different MTH engine to see if it will work. If it doesn't then I believe the chip gets a memory state set that doesn't allow it to work with Original MTH Processor firmware/software, but it will work with QSI systems that are capable of select/deselect.

It would seem to me that since these boards are the same, that QSI could have made a repair chip to reallow the original chip and processor to work. Given that there are no more chips to be had, other than what MTH has in stock, it would be nice to salvage the original chips.

Will update this after my additional test. I have several other deselected chips I am going to try in the Weaver QS-2+ system. G

Updated for test with MTH Engine. Well the original deselected chip ran fine in my MTH J that is not on the list. I ran it through all functions and did a reset #18.

I then put my J chip in the deselected engine. It worked fine including getting into a reset #18.

So then I put the original chip in the original engine and it still would not work.

Don't know what conclusion to draw other that the combination won't work once deselected. Don't know if there is a type of hard reset for the MTH PS-1 processor. Resetting Defaults won't do it. I know that the Q3000 and some QSI have this capability, not sure if there was a procedure for the MTH version. G
Last edited by GGG
G GGGOGR Forum Member

Well I finnaly solved this. I posted a procedure on the MTH DCS forum below for fixing a PS-1 engine.

While I finally received some info from TinScale it was cryptic, but researching some of the chips on the board and piecing together the syptoms and the information I was provided this is how the PS-1 boards work.

There is a MICROPROCESSOR that executes the software and sound files that are on the...

PROM which is a Read Only Memory chip. This is the removeable chip that sits in the PLCC fixture.

The previous unknow device is a small 8 pin dip chip which is an EEPROM and watchdog circuit that sits near the PROM. This has 2K of non-voliatale memory and stores the selectable features and other state changes during the shutdown sequence. This memory is used during startup also.

For the engines on the MTH list, the software on the PROM most likely can not bypass any established ENG ID the processor stores on the EEPROM. During a shutdown with a dead battery, apparently an ENG ID is established and this prevents the engine from starting up. This is a QSI feature that was used to allow an engine to remain on the track with voltage yet not respond to changes in voltage and direction, until addressed by it's ENG ID (a sequence of horn presses).

With a good "QSI" chip that works on MTH PS-1, you can start the engine up, and then cycle through the reset states until you get position #3 and #17 and clear the ENG ID. (Or you purchase the newer version of your chip from MTH which has software to avoid the ENG ID issue.  This is the permanent fix).  This clears the EEPROM memory for ENG ID, followed by a RESET #18. ENG ID does not have a default setting, so that is why a reset #18 doesn't work alone.

This explains why an engine that won't work with it's original chip after being deselected will work with a different chip not on the list (Software skips over the ENG ID issue) and the bad chip works in another engine (That EEPROM has not stored an ENGINE ID).

Hope this helps folks with deselected PS-1. G

Last edited by GGG
Views: 2,190
Post

Membership Required

We're sorry. You must be signed in to continue. Sign In or Register
×
Are you sure you want to remove from your Block List?
×
Loading...
×
When you block a person, they can no longer invite you to a private message or post to your profile wall. Replies and comments they make will be collapsed/hidden by default. Finally, you'll never receive email notifications about content they create or likes they designate for your content.
Note: if you proceed, you will no longer be following .
×
Link copied to your clipboard.
×

Access to this requires an OGR Forum Supporting Membership

OGR Forum Supporting Membership

Help support this forum with an OGR Forum Supporting Membership.
You will be able to watch the videos in the INSTRUCTIONAL VIDEO FORUM!
A one-year OGR Forum Supporting Membership is only $12 per year, so sign up now!


    OR     


Access the ALL the OGR VIDEO FORUMS AND  over 300 back issues of OGR
with a DIGITAL SUBSCRIPTION!



$12.00 per 12 Months (plus tax if applicable)
×