Home brew demo Board

[Mike Cook]

I would like to go ahead and start constructing a home made demo board to use with the Propeller while I wait for the Propeller to ship. I am starting with the schematic posted here:

http://forums.parallax.com/attachmen...chmentid=41055

and will be adding the composite video out circuit as shown below:

For color composite video out you need three pins, three resistors, and an RCA jack: 
 
A12 -----[ 1.1k ]----*-----> composite out
                     |
A13 -----[ 560 ]-----*
                     |
A14 -----[ 270 ]-----+

I have a few questions for Parallax or anyone that presently has a demo board:

1. I am planning on using a 5MHz crystal, what type of crystal is needed for the Propeller? Load Capacitance: specified or series?

2. The resistors for the composite circuit, do they need to be 1% or will 5% due.

3. The pin out the composite circuit, is this the default for the TV.SPIN object.

Thanks,

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

[Mike Cook]

OK, I think I found the answer for my crystal question, posted here in Ken's *.PDF, last page:

http://forums.parallax.com/showthread.php?p=572669

Note that no external resistors or capacitors are required for crystals and resonators.
 
OSCM1   OSCM0   XOUT Resistance   XIN/XOUT Capacitance   Frequency Range
  0       1        2000 Ohms             36pF              4MHz to 16MHz Crystal/Resonator

So am I correct in assuming that I can use a crystal with a specified capacitance of 20pF as long as I don't exceed 36pF? This is the one I'm thinking about ordering:

http://www.ecsxtal.com/pdf3/HC_49US.pdf

http://www.mouser.com/index.cfm?hand..._pcodeid=52003

Thanks,


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

[Beau Schwabe (Parallax)]

Mike,

If you are interested in doing video, the cut of the crystal does matter.

I will try to find out the difference and let you know.

You can compensate for the crystal cut in software by adding an offset of 3000Hz to the actual
crystal frequency you are using and specifying the '_xinfreq=' value in the IDE with that offset.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

[Mike Cook]

Hopefully the schematic that Ken posted here:

http://forums.parallax.com/attachmen...chmentid=41055

is correct!··

Here's what's left to do....

1. USB2SER & 24LC256 will be here Tuesday April 4, 2006 + 2 types of 5 MHz xtals, (20pF and series)

2. 3.3 volt PS works and I've electrically checked the board verses the schematic

3. Need to add a +5VDC supply

4. Mount this board to a chunk of LEXAN

5. Add 2 RS 276-174 plastic proto boards

6. Mount the Matrix Orbital LCD2041 serial display



·

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

Post Edited (Mike Cook) : 4/2/2006 7:07:05 PM GMT

[Jon Williams]

Do you have pull-ups on the SDA and SCL lines of the 24LC256 (connect them to Vdd, not the RST line)? Even through the Propeller drives the SCL line when loading it's program, you may want to put a pull-up on this line to work with the I2C object I'm writing -- that can use those same pins.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax

[Mike Cook]

Jon,

No I do not have pull up's in the SDA & SCL lines, thought it was strange that they did not have these. I followed the schematic that Ken Gracey posted here:

http://forums.parallax.com/attachmen...chmentid=41055


I usually use pull up's on AVRs, but I was just following the schematic that Ken Gracey posted.

What you would you suggest 4.7K's?

Is there a problem with the posted schematic?

The schematic that·Ken Gracey posted seems to be the same·that was posted·on this forum and published in the April edition of Nuts & Volts.

Thanks,




▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

Post Edited (Mike Cook) : 4/2/2006 7:39:14 PM GMT

[Jon Williams]

4.7k is fine. Our demo board has a pull-up on the SDA only because it drives the SCL pin -- but if you write an I2C object that doesn't do that you'll need the pull-up; it's cheap, you might as well do it. And please trust me that they should be connected to Vdd of the board's supply, not RST of the USB2SER (it won't hurt, it's just that when the USB2SER is gone, so is the voltage for your pull-up(s).

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax

[Mike Cook]

So if I am understanding you correctly I should·include the 10K between the SDA and the RST on the USB2SER·in the original schematic and add 4.7k's to SDA & SCL to VDD?

Like this? 10k from USB2SER· to SDA (Propeller pin 38) is under the ZIF socket.

·

Thanks,


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

Post Edited (Mike Cook) : 4/2/2006 8:04:17 PM GMT

[Jon Williams]

No; remove the 10K between RST and SDA. Connect SDA to 3.3v through a 4.7K (standard I2C pull-up); same for the SCL line. The schematic shown will only work if the USB2SER is plugged in -- should you test in stand-alone mode without the USB2SER (where the program is loaded from the EEPROM) it won't work because the SDA pull-up will have no power so you'll get nothing but zeros from the EE.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax

[Mike Cook]

OK, deleting (cutting destroying the 10K between the USB2SER (RES)·to PIN 38 Propeller SDA)·as shown on the the propeller schematic posted here:

http://forums.parallax.com/attachmen...chmentid=41055

SDA & SCL are now pulled up to +3.3 via a 4.7K resistor on each line.

So what I gather form this conversation is that there is an ERROR in the original schematic as posted here:

http://forums.parallax.com/attachmen...chmentid=41055

and in the April 2006 Nuts & Volts Atricle. The RES from the USB2SER should only go to PIN 11 of the propeller chip (RESn)·and NOT be connected via a 10K series resistor to pin 38 (Propeller SDA) for stand alone operation. Is there any harm in in including it?. Is it needed for programming the Propeller via the IDE?

Thanks Jon!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

Post Edited (Mike Cook) : 4/2/2006 9:16:40 PM GMT

[Brian Riley]

What is the word on a crystal? What frequency do we want and what cut? I have a bunch here 2, 4, 8, 20, and 50 MHz series cut ~20pf ... can I use any of them? I have resonators at 4, 8, 20, and 50 MHz. any of them?

cheers ... BBR

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
cheers ... brian riley, n1bq, underhill center, vermont

[Jon Williams]

Let me try again:·That circuit will work -- but only if you leave the USB2SER plugged in and connected to your computer.· Since the USB2SER is a programming device and won't always remain connected, using its RST line (normally high) to pull-up the SDA is not really a great idea.· I have sent a note to the person who drew the schematic to get it corrected in a manner that will work with or without the USB2SER.· Keep things simple, use your 3.3v supply for the SDA and SCL pull-ups.· The only thing that the USB2SER.RST line should connect to is the Propellers RSTn pin.· BTW, if you want a reset button on your board, connect a n.o. pushbutton between RESn and Vss.

Mike Cook said...
OK, deleting (cutting destroying the 10K between the USB2SER (RES)·to PIN 38 Propeller SDA)·as shown on the the propeller schematic posted here:

http://forums.parallax.com/attachmen...chmentid=41055

SDA & SCL are now pulled up to +3.3 via a 4.7K resistor on each line.

So what I gather form this conversation is that there is an ERROR in the original schematic as posted here:

http://forums.parallax.com/attachmen...chmentid=41055

and in the April 2006 Nuts & Volts Atricle. The RES from the USB2SER should only go to PIN 11 of the propeller chip (RESn)·and NOT be connected via a 10K series resistor to pin 38 (Propeller SDA) for stand alone operation. Is there any harm in in including it?. Is it needed for programming the Propeller via the IDE?

Thanks Jon!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax

[Mike Cook]

OK, Understood, the posted schematic will work as long as the USB2SER stays connected.

So my last changes to the curcuit of deleting the 10k between the USB2SER.RES to the Propeller.PIN38 (SDA) and adding the 4.7K pull ups to SDA and SCL on the 24LC256 to +3.3VDC·seem to be correct.

With the USB2SER disconnected, what state should the Propeller PIN11 (RESn) be? Should it be left floating or tied to +3.3 VDC with a weak pull up such as a 10K·or 100K?

I'll probably add the reset button to the circuit next.

Thanks Jon,






▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

Post Edited (Mike Cook) : 4/3/2006 1:56:04 AM GMT

[Beau Schwabe (Parallax)]

Brian,

"What is the word on a crystal?"

I still need to find out.... Only reason I bring this up, is that early on I was using a 10MHz out of a junk box, and
the question came up with Chip as to why I was only seeing black and white on my TV monitor. He mentioned
something about the crystal overtone messing with the synthesized color burst. I forget now if the crystal should
be series cut or parallel cut. Perhaps Chip or Jeff can help to answer this question.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

[cgracey]

Hold on! This is not true. This works perfectly without the USB2SER plugged in. This was quite intentional and not a mistake. This is optimal for power consumption since during brown-outs, current is not going to be sourced into the EEPROM's SDA line when the EEPROM is potentially pulling it low.
With·BOEn·wired low, the RESn pin will pull high via an internal 5k resistor·as long as·VDD is above ~2.6V. If VDD drops below this threshold, the RESn pin will pull low via the same 5k resistor and hold the Propeller in reset mode. Because the RESn input is a Schmitt trigger, there is no problem using it as a pull-up source for SDA, as long as an additional resistor is used. A 10k works fine.
The USB2SER and the Propeller Clip have only·a low-pulsing NPN transistor connected to their reset·output, and no pull-up, at all. The Propeller is doing the pull-up internally.
What Jon's mainly concerned about is being able to use the existing I2C·connection for other I2C peripherals, not just the 24LC256. For such an application, two separate resistors on SDA and SCL to VDD is the way to go. The one-resistor-to-RESn is a minimal approach when only the EEPROM is connected and you want to keep things simple and lowest-power.
Of course, you could connect I2C devices to ANY of the Propeller's pins. Jon would like to not waste another 2 pins, though, since 28 and 29 are already being used for I2C.

Jon Williams (Parallax) said...
No; remove the 10K between RST and SDA. Connect SDA to 3.3v through a 4.7K (standard I2C pull-up); same for the SCL line. The schematic shown will only work if the USB2SER is plugged in -- should you test in stand-alone mode without the USB2SER (where the program is loaded from the EEPROM) it won't work because the SDA pull-up will have no power so you'll get nothing but zeros from the EE.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔


Chip Gracey
Parallax, Inc.

Post Edited (Chip Gracey) : 4/3/2006 4:33:11 AM GMT

[Mike Cook]

Thanks Chip for clearing this up.

I'll put the 10K back in the circuit and leave the additional 4.7K pull ups in on the SDA & SCL lines so I can use the I2C bus for other I2C chips.

Mike

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

[cgracey]

Well, you might as well·get rid of the 10k from·SDA to RESn·if you'll also have a 4.7k from SDA to VDD. I think for your case, 4.7k resistors from SDA and SCL to VDD would be optimal. The alternative would be to have, perhaps, 20k resistors from SDA and SCL to RESn, but that's getting to be fairly high-impedance (you don't want to over-load RESn, as it IS a reset input, though Schmitt triggered, with a 5k output impedance).

Mike Cook said...

Thanks Chip for clearing this up.

I'll put the 10K back in the circuit and leave the additional 4.7K pull ups in on the SDA & SCL lines so I can use the I2C bus for other I2C chips.

Mike

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔


Chip Gracey
Parallax, Inc.

[Mike Cook]

Chip,

Good timing! I was heating up the soldering iron! I'll leave the 10K out, I had not had a chance to install it yet.

Thanks to ALL·for the help!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike

[Jon Williams]

Sorry, Mike, I've been so focused on developing Spin objects that I hadn't explored all the details of the RESn and BOEn lines.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax

[Mike Cook]

Jon,

No problem at all! There's always going to be a few "bumps in the road" when working with a new chip. In fact if you look to the right of the 24LC256 chip, in the above post, I'll probably put a 4 pin header there so I easily can extend the I2C bus out to the bread board or elsewhere.

Thanks Again,

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Mike