looking for a quick sanity check

[cncmachineguy]

After 10+ years of playing with PIC's, Its time to talk to a puter. I have attached my test schematic for critique. Please note all the optional pin names may not be correct on the PIC, I tend to get rather lazy about changing all the names. But I have connected the correct pins for ICSP and Tx2/Rx2.

Heres my setup I am thinking about:
Laptop running win7-64
USB->RS232 cable
ICD3->ICSP
Power circuit from ICD3 through ICSP for serial testing
Some sort of hyperterminal style PC side (prolly Daves (Mackrackit) serial.py code)
Hardware USART PIC side
PBP 2.6a
Int clock

Now I would love to use USB instead of RS232, but that seems like trying to run when I can't even walk yet. So I guess my question is will the above collection of parts along with the proposed circuit give me the ability to make this work?

4938

[mackrackit]

I am used to using the PicKit2 so the programming header looks off, what are you using?

I would not tie the extra pins on the DB9 together, just use 2,3, and 5 as you have them.

[cncmachineguy]

thanks Dave. I use ICD2 or 3 So all my ICSP headers are custom. I do think I will configure it to be the same as Pickit2 as it seems to be populer.
I will remove the jumpers, that came from the 3311 datasheet.

Any thoughts on the usb->232 cable? will this work?

[mackrackit]

I must have a different data sheet, I do not see the jumpers. But I am still half asleep...
http://datasheets.maxim-ic.com/en/ds/MAX3311-MAX3313.pdf

Are you talking one of those USB to serial conversion cables? If so, yes. Those conversion cables will also work if you are using SERIN2/SEROUT2 inverted without a MAX chip.

[dhouston]

One thing to be aware of is that USB-serial adapters require conversions on both ends, serial-to-USB & USB-to-serial. This can be a factor in time critical applications. IOW, 9600bps will not be a true measure of throughput. I have encountered problems with some bootloaders when using USB-serial and, also, with ethernet-serial adapters. Also, while not a factor here, some USB adapters do not handle all of the modem control lines like DTR, etc.

[cncmachineguy]

I must have a different data sheet, I do not see the jumpers. But I am still half asleep...
http://datasheets.maxim-ic.com/en/ds/MAX3311-MAX3313.pdf

No, I am the one asleep, after looking at the datasheet again, I realized that came from some circuit I found on the net.


Are you talking one of those USB to serial conversion cables? If so, yes. Those conversion cables will also work if you are using SERIN2/SEROUT2 inverted without a MAX chip.
Is this Yes it will work as I want, and also it will work using SERIN2/SEROUT2...?
The target puter for this app has an actual rs232 port, but I am wanting to play with my laptop first.

[cncmachineguy]

One thing to be aware of is that USB-serial adapters require conversions on both ends, serial-to-USB & USB-to-serial.

I'm not sure what you mean here :(

[mackrackit]

I use these with ny laptops.
http://www.tripplite.com/en/products/model.cfm?txtModelID=2430
As long as the signal is inverted, either by software or hardware it will work.

[dhouston]

I'm not sure what you mean here :(I doubt it's a factor here but...

On the PC end your serial signal gets converted to USB before it's transmitted. When received at the adapter it is converted from USB back to serial and then passed to the PIC. If your link is bidirectional the same conversions take place on signals from your PIC to your PC. At 9600bps, it's not much of an issue but at higher bit rates it can be significant. I've seen firmware downloads to an MCU that took 5 or more times as long over a USB-serial-adapter vs. a straight serial link. In some cases the download failed because the supervisory software didn't receive an ACK within the expected time. This can be even more pronounced over ethernet-serial adapters if there is heavy network traffic.

As I noted, it's just something to be aware of so you won't be scratching your head should you encounter it.

[dhouston]

I use these with ny laptops.
http://www.tripplite.com/en/products/model.cfm?txtModelID=2430I like these...http://www.byterunner.com/byterunner/product_name=Y-105/user-id=/password=/exchange=/exact_match=exact. They work with Windows, Linux and OSX. You need drivers for Windows & OSX.

[mackrackit]

I like these...http://www.byterunner.com/byterunner/product_name=Y-105/user-id=/password=/exchange=/exact_match=exact. They work with Windows, Linux and OSX. You need drivers for Windows & OSX.
About half the cost too.
Thanks!!!

[cncmachineguy]

Well maybe I am, but I don't know if I know more now or less. Heres where I have stumped myself.

Now the 3311 inverts the signal, so should I not use it if I am using the 232 to usb? Or does the cable take care of this? I would stupidly expect the 232 to usb cable to send/receive proper usb signals on the usb side and proper 232 signals on the 232 side.
This sounds like daveH post.

[mackrackit]

When going to a PC, the serial signal needs to leave the hardware inverted(PIC and components). Does not matter if you are going to a regular serial port or using a USB to serial converter.

The PIC, if using the hardware ports "HSEROUT" sends a TRUE signal so the MAX is needed for the inversion.
If bit-banging, "SEROUT2", then you can invert in PIC software.

[cncmachineguy]

I think I have enough to get in trouble with now. I am off to put my test circuit together and start talking (or not).