Legacy and gargraves remote switches

The DZ-2001 is made by Z-Stuff and sold by Gargraves.

I use DZ-2500 switch machines and they work well, but they are a bit temperamental and take a bit to adjust.  Don’t need the DZ-2001 when using DZ-2500s since the DZ-2500s have built in TMCC.

I have to disagree with Ron.

DZ-2500's do need a serial feed coming from one of two places: a DZ-2001 or a CSM2.  I would recommend sticking with the DZ-2001 serial feed exclusively as the CSM2 serial connection is not predictable.  GunRunnerJohn and I have been struggling with DZ-2500's for almost 10 years and the final verdict is very much DZ-2500+DZ-2001.

In deciding between the DZ-2001A or DZ-2001L, the DZ-2001L is the latest version and would be the best choice.

A little more info, the DZ-2500 does not understand anything directly from the TMCC base and requires an Opto-Isolated Serial Data line that the DZ-2001 provides to work.  Lionel attempted to replicate this serial data line in the CSM2 but it doesn't always work and you wind up chasing your tail trying to debug it.

I think GRJ would agree completely with my assessment.

@LionelAG posted:

I think GRJ would agree completely with my assessment.

I have actually diagnosed what I believe is going on with the CSM2 serial data, but I haven't bothered to switch from the DZ-2001 since it's working fine.

The issue with the CSM2 serial data and the DZ-2500 switch machines freezing on power-up appears to be the timing of the initialization of the CSM2 serial data output.  It appears there is some uncertainty of the serial output from the CSM2 for a brief period of time when everything is initializing during power up.  This randomly confuses the DZ-2500 initialization and causes them to be in manual mode instead of command mode.  The cure is to power on the CSM2 a couple seconds before the DZ-2500 and everything appears to work fine.  When I manually powered the DZ-2500's after the CSM2 had previously powered up, I never got the lock-out.  If I power the CSM2 and DZ-2500's up at the same time about 20-30% of the time, many or all of the DZ-2500's are in La La land.

If my lazy streak ever gets cured, I'll add in the delay on powering the DZ-2500's to allow the CSM2 serial data to stabilize and use them again, but I have lots of other fish to fry before I spend time on that issue.

@gunrunnerjohn posted:

I have actually diagnosed what I believe is going on with the CSM2 serial data, but I haven't bothered to switch from the DZ-2001 since it's working fine.

The issue with the CSM2 serial data and the DZ-2500 switch machines freezing on power-up appears to be the timing of the initialization of the CSM2 serial data output.  It appears there is some uncertainty of the serial output from the CSM2 for a brief period of time when everything is initializing during power up.  This randomly confuses the DZ-2500 initialization and causes them to be in manual mode instead of command mode.  The cure is to power on the CSM2 a couple seconds before the DZ-2500 and everything appears to work fine.  When I manually powered the DZ-2500's after the CSM2 had previously powered up, I never got the lock-out.  If I power the CSM2 and DZ-2500's up at the same time about 20-30% of the time, many or all of the DZ-2500's are in La La land.

If my lazy streak ever gets cured, I'll add in the delay on powering the DZ-2500's to allow the CSM2 serial data to stabilize and use them again, but I have lots of other fish to fry before I spend time on that issue.

Thanks for the update on your diagnostic activities.  Please let me know if you make any further progress on this as the CSM2 is a nice package by itself.

I still use the CSM2, just not it's serial outputs.

@gunrunnerjohn posted:

I still use the CSM2, just not it's serial outputs.

Same here.

So I have received everything and have it wired according to the supplied instructions. Red on red, black on black, and the white data wire on the DZ-2001L to the blue data wire on the DZ-2500. The switch powers up, and works when I push the button on the DZ-2500 for manual operation. however, I can’t get it into programming mode to set an ID. The instructions say to hold the button on the switch motor for four seconds and release when both the green and red leds start flashing but they never start flashing.

Some Good news! Everything works almost as it should. Turns out the supplied instructions are not entirely accurate. You have to hold down the button on the switch motor for way more than four seconds, probably closer to ten, before both the red and green LEDs start to flash and is ready to be programmed.

I can see what people mean when they say the switch motor is a bit temperamental. I was testing it out, following the instructions, exploring how the switch operates and I tried to change which light is lit when the switch is aligned for the left turnout, mind you this is a WYE switch. Currently, when the switch is aligned to the left, the red LED is lit. I wanted to change this so that the green LED is lit because the left turn continues on around the loop, whereas a right turn branches off to a siding. So I follow the directions and program the switch so that the green light is the one that is lit when the switch is aligned to the left, but afterwards, when I try to throw the switch to the right, the motor does not stay in place. It's as if the motor doesn't have the strength to hold back the force of the spring, it just slides back to the left. It's strange that it only does it when I try to program the switch so that the green LED is lit when it is thrown to the left. If it is programmed the opposite way, so that the red LED is lit when the switch is thrown to the left, it operates just like it should and the motor stays in place when it is thrown to the right.

Not the best experience so far. I hope not every switch is like this.

Your experience above sounds like the position of the switch may need to be adjusted on the physical Gargraves switch.  The machines won't force a throw if something is holding it back.  Otherwise, I don't know why changing the light indicator would have anything to do with the machine not completing the throw if it did so before.