Reward Details:
I'm currently offering a $50 paypal reward (likely to go up as time progresses). Your solution should follow procedures outlined in the MSC-400 reference manual, but any work around that I like will result in full payment of said reward. Your solution should be reproducible on my equipment (not just yours), although if you make a video showing it works on your equipment (or otherwise show my equipment is faulty) you'll likely still get the reward. You must use an MSC-400, MX-900 and a PC as outlined under "Preferred ways to reproduce this issue." I can send you the MSC-400 and MX-900 files as necessary (not the programs obviously, just my files).

Problem:
If I set up a custom RS232 device as outlined under "Preferred ways to reproduce the issue" (see below), RS232 commands are skipped, resulting is 1 out of 5 or 2 out of 10 failures when using an MX-900 and an MSC-400.

What I've done to try to solve the issue:
1. I reset the MSC-400 using the small button on the bottom. I've reset the MX-900 using it's menu. I then loaded completely new files.
2. I've checked the red receive light on the RFX-250 to ensure no interference.
3. I've tried two different MSC-400/RFX-250 combo's, (one from 2006, one from 2007). I'm not sure if this issue is revision related.
4. I've tried two different MX-900 remotes (one from 2007, one from 2008). I'm not sure if this issue is model specific.
5. I've checked the serial cable with a volt-ohm meter and have ensured that TX from the MSC-400 goes to RX on my test PC (running Hyperterminal) and that ground goes to ground. I also checked that RX went to TX on the PC.
6. I've physically moved the RFX-250 and MSC-400 (just in case it is interference which the light indicates its not)
7. I've verified that the button is being press all the way each time on the MX-900 as I enabled a press beep
8. I've checked nearly all hard and soft buttons on both MX-900s, they all result in skipped RS232 transmissions.
9. I'm using hyperterminal so I have an accurate count of good packet transmissions.

Temporary solution I've come up with that isn't good enough:
Create two smart macro groups from the same custom rs232 device in the MSC400 editor. Use the MX900 editor to toggle between these two smart macro triggers that send the same command by making a macro of variable type and inverting a variable called toggle on each press. Hyperteriminal shows this gives about 98% command accuracy. This isn't a good solution as it's limited to only one type of macro under the MX-900 editor (variable type). I'd like to be able to use the "on press hold" macro type with similar accuracy, and setting a toggle up each time I want to use RS232 is added work.

Preferred ways to reproduce the issue:
1. Use the latest MX900 editor and the latest MSC400 editor (goes without saying to use the same equipment as me)
2. Use your favorite terminal client to test the accuracy of the output (as outlined in the MSC-400 reference manual).
3. Ensure you set the MSC-400 up with at least one custom serial device. Define a couple commands that send test strings like "test" with a carriage return. Right click on the device so that the MSC-400 editor will auto generate a smart macro group for you. Bind the relevant smart macro trigger to a button using the MX-900 editor and the Universal Browser and use the "Normal" macro type.
4. Press the same button repeatedly every .75-1 seconds (that will "trigger" a smart macro to send the RS232 command), count the number of good transmission using the terminal client.

sounds like a memory cache'ing type of issue, where the unit (or any device in the system) is still processing the previous command when the next one comes in...explaining what you're trying to do/accomplish would be a better way to get responses and results.

ie, if you are trying to repeat a volume (or navigation or channel) up/down/left/right type of command say this is what you're trying to do, and list the device/model number of the device you're trying to control.

it could be that your control system is fine but your end device can not process 'x' number of inputs/commands in 'n' time period. You've done well to try and give us lots of information, but without the right information its hard to help.

Thanks Bryan.

You are right to suspect what I'm trying to control. We think alike, that's why I took that out of the equation and started troubleshooting with Hyperterminal.

Hyperterminal is a windows program that allows you to examine RS232 output from the MSC-400 (or any socket connection for that matter). There is no way that Hyperterminal could be too slow to react (and I've checked the port settings). With Hyperterminal, I know for sure what is being sent out by the MSC-400. I know for sure the work around from the original post works (but only for variable macros), but sending the same trigger on each button press doesn't.

I'm really hoping someone can verify the screwy results I'm having. It could be there is some setting I have wrong in the software. It may make more sense for me to post a test file, but I think there's enough info for someone with similar equipment to reproduce the issue. If you were to post a file that offers reliable custom rs232 commands, that'd obviously fix my problem as the only goal (for now) is to get the proper resone in Hyperterminal.

What I want do is have reliable button presses if I a user presses the same key more than once. It's not practical for every button to be a ramp button and use the press/hold work around and I don't think it would fix the issue I'm having anyways.

To be clear, show me that the MSC-400 and MX-900 can operate as designed using a terminal client (Hyperterminal, RealTerm etc...) for single button presses, send me test files that prove this with a few custom RS232 commands you'll win the reward :)

a test file may help...and exact equipment is what would be neccessary to duplicate the issue, unless you're saying that you've taken the equipment out of the scenario and are just viewing the strings as they come through via hyperterminal. If this is the case it leads me to believe its still a cacheing type of issue I was originally referring to (but understanding that the cacheing issue could be in the remote, the antenna or the msc) and would just be a function/limitation of the product.

My next question would be what brand and model number of receiver are you using, and can you do a repeat while pressed/held type of functionality and/or is there a ramp function? does IR work? use that instead.

Correct, I'm only connecting the MSC-400 to my PC and Hyperterminal, there by taking the device out of the equation. I'm not going any further until I get things to work in Hyperterminal ;)  All that is needed for the reward is to get things working in Hyperterminal.  This way we both can effectively use the same test device.

I'll worry about tailoring things to the specific device (which is a complex Home Automation system interfacing with HVAC, an alarm system, lighting and many media devices) after it is proven that Hyperterminal sees the correct output from the MSC-400. Given this RS232 is a must for communication between the MSC-400 and the home auatomation system.

Single IR commands sent to the MSC-400 don't exhibit the issue, only smart macro triggers to send text out an RS232 port. Press/hold wouldn't work as I need single press functionality to work 100% of the time; an example would be toggling speeds for fast forward and reverse. If a user hits the reverse button 4 times to get the fastest reverse speed for a component, one RS232 command may drop.

If I use press/hold there's no way I can see to toggle the smart macro sent out for a single press since the MX-900 editor is limited compared to the other more advanced remotes. Without toggling the smart macro sent out in between presses, RS232 reliability goes out the window...

The caching idea of yours sounds very plausible, except why hasn't anyone else complained of this issue before? I called URC about the issue three times and they were useless...

I'll see about making a simple CCP test file if you have an MSC-400 and an MX-900 to try?  I'll use my toggle work around for the up arrow, and leave the down arrow alone...  Hopefully someone else can then reproduce this issue. I think this is the first step, proving that a completely different system has the same issue.

EDIT:

Forgot to mention that repeat buttons have an issue documented in other threads. They always give an extra command. There's the press hold work around, but it doesn't fully work. If you look at the RS232 output in Hyperterminal, you see that you always end up with an extra command on any hold operation. Kind of annoying.

Here's a sample CCP file showing what I'm saying. Use the default serial settings for the MSC400 when setting up Hyperterminal.

UP - Normal Macro that sends a smart macro trigger to send TEST out of serial port 12 of the MSC-400. When you press Up 10 times, you should see at least 2/10 commands that are lost using Hyperterminal.

DOWN - Uses a toggle method (variable macro) to switch smart macro triggers on each button press. This ensures no single smart macro trigger is sent twice in a row, but results in the same ascii text being sent each time on port 12. When you press DOWN 10 times, RS232 dropped commands diminish and the button actually works.

LEFT and RIGHT illustrate the repeat issue. RIGHT uses the press/hold work around. It seems at random, one extra RS232 command is sent after the RIGHT button is released. The LEFT button almost always sends two commands on a single press.

Use the google document link below to download this test file...
docs.google.com/leaf

can't help any further as I don't/haven't had to use an msc-400 yet (for complex installs I use RTI) and I've never been a big fan of the mx900. I use the 450 and 980 with mrf260/350's and IR commands, and if more is needed I change lines.

good luck

Thanks for taking a look :)

I think you are dead on that it's some type of issue with how the MSC-400 and MX-900 cache things, but it'd be nice to know someone with a similar set up had the same issues, even on the latest hardware revision of the MSC-400. I'm hoping someone else can chime in so I'm going to raise the reward to $35...

I'm really confused why I didn't get more responses... No one uses the MSC-400 with an MX-900 for custom RS232 commands? I hope my free gift didn't offend anyone, I just trying to give back to those who spend a lot of time helping others on here...

It'd be helpful is some of the pro installers could just comment on what their experiences are with RS232 command reliability on the MSC-400 and MX-900... Maybe I'm not the only one and I should abandon the idea or live with the work around I developed?

i'd love to test out your program however i don't have a msc400 to test out here at the shop. i don't want to reply " it works, trust me i've done it before, my paypal is..." i personally have used the msc400 for several different avr's but never had to "analyze" the output because the database has always worked.

I use the MSC-400 with my MX-980. I have seen several very odd things (besides those mentioned already).
1) There is the 2 command thing (which by the way I have devices that actually work, then when I tried to fix the others, they all broke).
2) I commonly have issues if I try to send out a command more than once a second (single short taps on the button one second apart always work on mine, anything else - not so much).
3) The stupid IR problem (it puts out short burst of IR on ALL the ports before it sends the command you asked for on the correct port, and the MRF-350 does not do this)

If there were something else I could buy and program myself that was as nice as the pair, I'd be all over it, but I really love the MX-980 and RF control so I'm keeping them (and have purchased several)...

Interesting, this sounds similar to my issue. If I leave 3-5 seconds in between buttons, I have noticed RS232 reliability IS greatly improved, but still not perfect. You might try toggling smart macros using the method in the CCP file I posted above. It "fixes" this issue, but is a pain to do for all 50+ buttons I'm going to need to control my components, HVAC, alarm and lights over RS232...

I haven't noticed this bug yet. Does this occur on normal IR buttons or only on smart macros? I just tested this for a few IR buttons that output via the MSC-400 and only had output on the correct port. However, I only have things to output port 1 and just connected a spare blinkie to port 2. It may be you have to have other ports setup in the editor for it to show the bug. PS: I'm using a Xantech IR hub to connect everything so that I can also interface a GC-100 network based IR controller; this is why I only use port 1 for IR.

I agree. The MX900 remotes I have feel very nice and are made in Korea (not China). Of course, your remotes are even nicer ;)

If only URC support could do the same! The fact is, they even said the well known RS232 repeat twice issue didn't exist when I called them last week. I might try sending them my test file, but I'm not a dealer so they may not even try it for me.

I'm beginning to think this is a hardware limitation of the MX900 and/or MSC400. I've raised the award to $50 if someone can prove me wrong :)

If you do prove me wrong and I'm able to replace hardware to fix my issue (e.g. newer MSC400 or MX900), you'll definitely receive $50...!

Have you tried inserting time delay between commands? This may sound almost too easy, but it proved to help many times in past.