Help with RENARD 16 with Integrated SSRs

[MLITE]

Sorry for coming in at the last minute. I’m a new member and having a little trouble getting my setup to work. Unfortunately, I missed the window for ordering the Renard 16 with integrated SSRs. I have been using the following document to build my system - 16CPicDimmedSSR_Theory_of_Operation_Rev1.9. I used all the recommend parts specified in the BOM and followed the instructions exactly with the exception of the following.

At this point I am not using an external 18.432 MHz oscillator and I am not bringing power to the board from the PC by way of the CAT5 cable. I am just using a radio shack walwart. I have an RS232 to RS485 converter plugged into COM1 so I only used the 120ohm resistor based on the instructions in the PDF “If you are using RS485 for PC to controller then the only thing you need is the 120 ohm resister between the A/B leads on the 485.”

My setup -
1st breadboard for logic – this board contains the 5V reg, RS485 chip, & PIC

2nd board for 120V – this board contains 1 Opto-Triac, 1Triac, and ZC Opto (just to see if I can get it to work)

I’m using CAT5 cable. Only 1 pair is being used (blue & blue/white) since I am bringing power to the board separately.

Here is what is happening…
I hooked everything up and programmed the PIC with ren-diag-20070815a.asm with one small modification. I am not using an external 18.432 MHz oscillator. I simply used the alternate config code –

__CONFIG _FCMEN_OFF & _IESO_OFF & _BOD_OFF & _CPD_OFF & _CP_OFF & _MCLRE_OFF & _PWRTE_ON & _WDT_OFF & _INTOSCIO

Once powered up the heartbeat and ZC led blinked (not in unison, but I don’t believe they are intended to). I started the VIXEN software and the 'one-shot' rx_okay LED lit on and off periodically. Since I am only using one PIC at this time I had shorted pin 5 & 6 on the PIC as discussed in the Theory of Design document. However, I had to remove the wire in order for the 'one-shot' rx_okay LED work with the Diag code. I also pulled pin 2 low on the OptoTriac and the light strand lit up.

So at this point I’m feeling pretty good. Heartbeat, ZC, rx, and wiring of 120V all looks good. Next I remove the PIC and program it with renard-20070407c.asm. Once again I changed the config code and also had to change the define statement for baud rate to 19200. I plugged it in and started the software expecting to see the lights going to the music. Instead all I saw was complete darkness.

Things I checked –
Configured COM1 just about every way you can imagine.
Checked, double checked, and checked again that the firmware, VIXEN, and COM1 settings all matched.

Any help would be greatly appreciated.

[xmus]

Please attach pictures of this monster!

Is your wall wart 12V or 5V? Are you running it to TP2 or TP6?

My suggestions:

1) You may have the polarity of RS485 swapped
2) Your FW must have 19200 baud and INTOSC enabled (but you know that, but are you sure you did it correctly? ). Attach your ASM code for us to review. I see where you say you configured for INTOSC, but I don't see where you modified the code to run at 19200.

I guess you know your com port is setup good because of the RX receive light.....

[MLITE]

Thanks for the quick reply. I am amazed at the level and speed of the help that is provided by everyone at this site.

I will be sure to check the RS485 connection and baud rate. I will take some pics tonight when I get home from work and also send a copy of the code I am using. I will add labels to the pics in order to help you understand how I wired it. Sorry, it is not going to be pretty, but wanted to lay everything out prior to etching my own board. Also, it might be little bit later in the evening. Today is my B-day and I have a feeling my wife and kids are up to something.

Thanks again for the speedy response.

[P. Short]

I haven't read your comments in detail, but two things jump out (although you may have done them without report them). If you are using the internal oscillator, you need to change the line in the .asm file that says what the clock speed is...change it to 8000000. Also, you normally need to bring a ground wire down the RS485 cable as well as the two data lines.

--

Phil

[MLITE]

I am still working on the pictures of the boards, but do have some good news. Using the PWM firmware with the changes that Phil suggested provided light at the correct time. However the lights are flickering very badly and none of the effects work correctly. Also, the lights only work when I removed the wire shorting pin 5 & 6 of the PIC (since I am only using 1 PIC). The non-pwm firmware still doesn't work at all.

The wall wart I am using is 9V 210mA. I also tried configuring COM1 every way possible. I currently have it set at -

Data Bits = 8, Parity = none, Stop Bits = 1, Flow Control = none, Use FIFO buffers = unchecked.

I will upload the pictures this evening. Thanks again for all the help.

[xmus]

Also, the lights only work when I removed the wire shorting pin 5 & 6 of the PIC (since I am only using 1 PIC). The non-pwm firmware still doesn't work at all.

The wall wart I am using is 9V 210mA. I also tried configuring COM1 every way possible. I currently have it set at -

Data Bits = 8, Parity = none, Stop Bits = 1, Flow Control = none, Use FIFO buffers = unchecked.

I will upload the pictures this evening. Thanks again for all the help.

I'm going to need a picture of the REN16 fully wired.
1) which 1/2 are you hooking lights to (and placing a pic in)?
2) Black HOT AC 120VAC must be present on the RIGHT half, even if there is no PIC or LIGHTS on the RIGHT HALF. The ZC gets its HOT BLACK from the right side.
3) your neutral wire must be connected to "N" not "NC" (NC=no connect, N=Neutral)
4) What baud rate is the serial port set to
5) What buad rate do you have your FW set to?
6) Are you supplying 9V to TP1 or TP6? (You are using the 5V regulator correct? Have you verified with a VOM that 5VDC is present at TP6?)
7) Are you using RS232 or RS485? If you had your polarity reversed..... you might get some weird behavior (might not work at all).


Suggestion, if it is "Kind of" working, try rolling the FW to 19200 baud, internal OSC, and setting the renard plugin to 19200 baud.

[MLITE]

I will add the pictures tonight. However, please remember I am working off two breadboards and did not make it in time to buy one of the kits.

Based on the DIAG code the ZC signal appears to be fine. I also checked the signal using my parallel port and the Digitrace software. The signal spacing was ~8 milliseconds apart.

I am applying 9v to the logic board with the radio shack wall wart. I confimed that 5v is present coming out of the regulator. I have an RS232 to RS485 converter plugged into my COM1 port (just like the one shown in the PDF). I confirmed that the CAT5 wires were not reversed when plugged into the breadboard. I also confirmed the baud rate of 19200 in the firmware, VIXEN, and PC port settings. I swapped the PIC thinking it might have been faulty, but still no luck.

I am guessing at this point I must have wired something incorrectly (hopefully that can be identified when I upload the pictures), ore there is a faulty chip (ZC opto maybe), or something is off with the timing.

[xmus]

8mS is about right, I think your ZC is working, as the diag chip confirms for you.
You don't need an oscillator because you are using INTOSC 19200.

I would say you are having a serial issue.

Trace the serial signal from origin to destination.

[MLITE]

Well here it is. I included an excel file with a picture of the breadboards, an excel file showing what I did, and the firmware that I am using.

Thanks again for any help you can provide.

[xmus]

Phil has no windows products (and he is the expert on Renard) so If you would like Phil to be a participant in your debug, then I suggest posting stuff that isn't in Microsoft formats.

[xmus]

Looks like you are using a white opto-triac, I haven't had good luck with those dimming. Can you swap it out for the recommended opto-triac? Please tell us the exact part number of your opto.

[xmus]

Since I am only using one PIC at this time I had shorted pin 5 & 6 on the PIC as discussed in the Theory of Design document

You should only do that if you are using my PCB, on your board you should not short pin 5 &6!

[MLITE]

Thanks for taking a look. I think on the excel file I showed that I removed the wire shorting PIC pin 5 & 6. In fact, that was the only way I was able to get the lights to work.

I will convert the excel file into a PDF so Phil can take a look if he chooses (although I'm sure there are more interesting things he would rather be doing).

I confirmed that the serial data coming from the RS232-RS485 converter appeared to be good. Please note that I only compared the A & B signal and did not try to decode the actual data. I would assume that since the Diag code was working and neither of the error leds were lighting that the serial data is good.

The one area that I strayed from the schematic in the PDF (16CPicDimmedSSR_Theory_of_Operation_Rev1.9) was the RS485 chip. I ignored all the other hardware since I am using the RS232-RS485 converter and only used the 120 ohm resistor as specifie in the PDF. Could this cause a problem?

I converted the excel file to PDF and also included the parts I am using at the end of the doc.

[P. Short]

Are you using a solderless proto board? Some of them split the power and ground rails in the middle of the board. Also, what is the 1K resistor and the red blob next to the regulator circuitry? Is that a power indicator?

--

Phil

Never mind...I finally scrolled down and saw the photos.

Phil

[P. Short]

How closely does the photo resemble your present setup? In the photograph the white opto-isolator looks like it's in upside down, the black dot to indicate pin '1' appears to be in the lower left corner near the high-voltage side (although I can't be sure because the photo is slightly out of focus).

--

Phil

[MLITE]

Phil,

Chip orientation should be 1 for 1 when comparing the picture and excel file.

The locations are not exact, but I least wanted to make it easier on anyone that is willing help to see how I made the connections and identify where I may have connected something incorrectly.

V/r,
Matt

[MLITE]

I made some progress last night. I decided to try a higher value resistor for the ZC OPTO circuit. As I increased the value the random flickering diminished and eventually went away. Also, once the flickering was completely gone I was able to do some dimming. Levels below 50 didn't seem to work too well, and the flickering became visible again.

Does anyone have any idea why this would be happening?

[xmus]

I made some progress last night. I decided to try a higher value resistor for the ZC OPTO circuit. As I increased the value the random flickering diminished and eventually went away. Also, once the flickering was completely gone I was able to do some dimming. Levels below 50 didn't seem to work too well, and the flickering became visible again.

Does anyone have any idea why this would be happening?

There are 3 resistors associated with the ZC... which one did you change?
There are the two fatties that go between the 120VAC, then there is the other one...

[MLITE]

I increased the other one. I believe the design in the PDF called for 27K .

[MLITE]

Just curious, but are there any boards still available for sale. Is it possible to order more? I realize the quantity discount wouldn't be there, but I'm not too worried about that. I really don't want to buy an already assembled board (light-o-ram, what fun is that).

[xmus]

I increased the other one. I believe the design in the PDF called for 27K .

(no more PCB's available, but I am doing another co-op (look in the co-op section))

Phil, What do you think about him increasing the value of the 27K?