After that, swap the power supply with a good engine, but I would only do that once and while monitoring amperage. It is easy for a bad processor to take out the charging circuit if you don't know what to look for.
GGG, is it worth the risk, or would a better choice to put the board on a tech's test set?
After 1.5 hours of charging, I added the engine to my (extra) remote. It went into ID #1. I changed the ID to #25, not #23.
All functions were normal.
The INFO screen read battery OK, and the chronometer had advanced the correct time
I then cycled the power off and on.
All functions were normal, even though the ID was now #25. However now the INFO screen read battery low. The chronometer had advanced.
I repeated the power off-on cycle, with the same results as above. (Functions normal, INFO screen read battery low).
In other words the loco now behaves exactly as it should, other than the BCR status rapidly dropping from "OK" to "Low". This is in marked contrast with what was happening earlier in the day
However, I got suspicious because of this erratic behavior, so I measured the voltage on the BCR, with the BCR removed from the engine. The fastest I could do this was within three minutes after shut down. The voltage was 1.54 Volts. One point five four.
I then reconnected the BCR, and measured the voltage directly to the BCR, with the engine on. I measured the same 1.54 Volts. I would guess my charging circuit has been compromised. As Stan2004 said:
It could be, for reasons TBD, that the BCR and associated circuitry is on the edge. That is, this is a repaired board that had a short and never quite worked right after that short. So even though it plays the 8 sec shutdown sound, the BCR conks out just after that and can't supply the power to complete the parameter-saving function. This is just speculation of course.
It may have happened during that first charge when I tripped the Lock-On, or it may have been from the original short circuit when the engine was upside down that started this whole affair
Rather than doing any swaping of boards and components, I plan to give this to a qualified tech and ask them to replace the charging board.
Comments? And thanks again for everyone's help
What did you charge a battery or BCR for 1.5Hrs? Is this a new battery or an old one?
We are at the point of seeing 10-12 year old 2.4V batteries, and they will hold a surface charge, but not last for much operation. You need to use a new fully charged battery.
The initial battery OK is good, the later battery low is bad.
Disconnect battery and measure the voltage at the battery charge terminal on the engine. If you have a spare 2 pin molex wire you can plug it in to attach to VM leads. Just make sure you don't touch leads together
Turn engine on and measure voltage. Should be 5V DC (no load).
Turn engine off. Attach battery to connector and turn engine back on. With a good battery it should read 2.5 to 2.7V. If lower the battery is most likely bad.
Only way to properly test a battery is under a load. No load voltage can read normal, but the second it needs to supply current the voltage falls off fast.
If battery is good, time for a tech that knows boards and component level repairs to look at it. That would be me!
Since this is not taught at MTH and I am the guy that reverse engineered how the board functions. G
GGG
I guess I wasn't clear in my previous post, probably because it was part of a long string of posts
What did you charge a battery or BCR for 1.5Hrs? Is this a new battery or an old one?
BCR, I don't have a good 3 Volt battery
The initial battery OK is good, the later battery low is bad.
I am aware of that. The point I was making was it went from "OK" to "Bad" in the interval of one power off- power on cycle. (Power off means track power shut off)
Disconnect battery and measure the voltage at the battery charge terminal on the engine. If you have a spare 2 pin molex wire you can plug it in to attach to VM leads. Just make sure you don't touch leads together
with the BCR disconnected, I measure 5.15 Volts
Turn engine off. Attach battery to connector and turn engine back on. With a good battery it should read 2.5 to 2.7V. If lower the battery is most likely bad.
Again, all I have is a BCR. Per my post above, the voltage was 1.54 Volts. I am aware this implies a bad BCR, but this is the second diagnosed "bad" new BCR. (Both BCR's were brand new, provided by the MTH trained tech. He said he has never seen a bad BCR. )
If battery is good, time for a tech that knows boards and component level repairs to look at it.
Assuming a BCR should charge to the same level as a battery, one possible explanation is the charging board can not drive enough current to charge the BCR. Another explanation is both BCRs are indeed bad. I can swap in a third BCR and see what I get, or if you think there is something bad about BCR's in general for this application, I can try to find a 3V battery.
That would be me!
So where are you located?
Sounds like the BCR is not getting charged, but the circuit to connect it to power supply works.
Shame you don't have a battery as it can help verify the processor is working with respect to memory transfers, since the battery doesn't need to be charged, but a BCR does after each power cycle.
I am in VA, e-mail in profile. I would just need the board set. G
Definitely send it to GGG seeing as you guys are just about neighbors.
That you were able to get the address (and chronometer) to "stick" through a power cycle once is fortuitous. That tells me the more-expensive processor board which has the memory chip that saves this data is OK.
Since you're sending it in anyway it's probably not worth over-analyzing, but a 1.54 Volt BCR will not be able to run the 8-sec power down sounds and save settings to memory. I don't have the numbers handy but you can do the calculation of how much energy is needed to do this. IIRC a BCR would need to be charged up to 4.x Volts. With a working charging board, a BCR should get adequately charged in a minute or so. It's interesting you let it charge for 1.5 hours and it worked. So it could be the charging circuit is broken and even though it reads 5V available (nothing connected), when loaded with the BCR, it can't provide the commensurate current to fully charge the BCR in a reasonable time. In retrospect, it's interesting the repair shop (in spite of saying he never saw a bad BCR) replaced it and charged you $20 or whatever when you first took it in!
BTW, one reason to send it to GGG instead of your local guy is GGG can do component-level repair. My guess is your local guy would simply swap out the charging board. Nothing wrong with that but if GGG does find something at the component-level, it pays a dividend as it adds to the OGR collective knowledge via a future thread.
OK. I have a hypothesis of what is going on. It has the benefit of fitting all my observations;
The charging system in this GG1 is weak. For whatever reason it can not drive its full design current, even though it can produce its full voltage. It could have been this way for a very long time (years!). The reason I never noticed it in the past was I had a battery in it, and a battery retains its charge for a long period of time, and thus can be trickle charged over a long time. A BCR, on the other hand, does not keep its charge very long, and thus must be charged up very quickly. Thus, as soon as I switched to a BCR, the weakness reared its ugly head, as the weak charging circuit could not charge up the BCR requisite minute or so before start up.
Remember I reported this having erratic behavior yesterday. It was just a matter of how long I let the loco charge up the BCR. About 1.5 hours was adequate
Here is the new data that supports this hypothesis; (NOTE, TO PREVENT CONFUSION, THIS IS THE FIRST TIME I USED A STAND ALONE CHARGER, OR A BATTERY RATHER THAN A BCR)
Test: I put the BCR on the stand alone battery charger. No locomotive involved, just the charger to the BCR. After five minutes, I disconnected the BCR. It read 4.30 V. After three hours, it read 2.4 Volts. Recall that after being in the engine, the highest it got, even after 1.5 hours of charging, was only 1.54 Volts.
Conclusion: The BCR is OK
Test: I found an old 3 Volt battery. I know its not new, its at least seven years old, but its all I got. It read 2.0 Volts. I put it in the engine, and while the engine behaved normally through several power off on cycles, the remote indicated the battery was "LOW"
Conclusion: Nothing wrong with the ability to write to memory or processor memory transfers. GOOD
Test: I charged the old battery overnight, plugging it directly into the stand alone charger. Again, no locomotive in the circuit. After 8 hours it read 2.4 Volts. I put it in the GG1, and the GG1 performed as normal for at least a dozen power off-on cycles over two hours. The remote always read the battery condition as OK
Conclusion: A fully charged battery will keep its charge. No significant parasitic loads
Test (from yesterday) Locomotive charging circuit produces 5.15 Volts with no load, only 1.54 Volts with a BCR load
Conclusion: Locomotive charging circuit is weak. Something is preventing it from driving full current into the BCR, and probably the battery
OVERALL CONCLUSION: A battery can function normally with a weak charging circuit. A BCR cannot. Thus my hypothesis which begins this post
Let the criticism begin
Although I have installed supercapacitors in a few of my locos as the batteries hot the 8+ years mark, I have done so because as I overaccumulate locos, some don't get used very often. I am still not sold on supercapacitors (which term includes BCRs) and trhink batteries are fine.
One of the sharpest guys in O=gauge model railroading, one who has long experience is repairing them, is Marty Fitzhenry. He stated on the forum that he sends all his boards to GGG for repair. I can think of no higher endorsement. I would never trust a LHS with a DCS board issue.
GGG: If you diagnose the problem, please post the result.
It's refreshing to see this level of curiousity so I'm happy to engage. Not criticism but comments:
As before, I can't imagine how a BCR charged to 1.54 Volts can run the power-down functions/sounds and save any changed data (address, chronometer, etc.) to memory. Energy in a capacitor (as used in a BCR) is proportional to the squared-voltage. So if a charged to 4.3V (your example), that is about 8 times the energy of a 1.54V capacitor. Anecdotal evidence shows a BCR can only run one power-down sequence per full charge...after all it must be re-charged after each use. So if it's only charged to 1/8th its capacity (i.e., 1.54V), there's not enough gas in the tank. OTOH a charged battery can run maybe 100 shutdown sequences since its energy capacity is so much higher.
If you are able to safely measure the BCR voltage with the engine running, you should see the voltage climb, albeit slowly, with a "weak" charging circuit. After 1.5 hours it apparently reaches 4.x Volts providing enough to perform one power-down sequence. Perhaps measure it every 5 minutes or whatever or between trips to the cabinet to refill the scotch glass. If you are able to measure the BCR during a power-down, it will collapse very quickly during the 8 sec audio sequence down to (apparently) 1.5V in your case.
If the BCR voltage just sits steady for minutes at 1.5 Volts when the engine is powered then your charging circuit is not just "weak" but it's anemic. That means whatever current is flowing is being "leaked" rather than going to charging (increasing the voltage) the capacitor. This tells me you have some kind of schizophrenic circuit that sometimes provides enough current to very slowly charge the capacitor...but other times can barely pull it's own weight so to speak.
Again, all this is great fodder for discussion while sipping scotch...but aren't you just going to send it to GGG? Or are you now considering living with the situation (as you have been for years) and maybe putting in a battery?
OVERALL CONCLUSION: A battery can function normally with a weak charging circuit. A BCR cannot. Thus my hypothesis which begins this post
Let the criticism begin
Great thread, especially like the scotch part. I'm glad to hear you were to change the ID and have it stick. I was curious whether the engine dropped into conventional at the slightest track power interruption?
Access to this requires an OGR Forum Supporting Membership
