PropStick and Professional Development Board

SSteve · 2006-05-04 22:27

Looking at the PropStick pictures makes it look like the DB9 port needs to hang down the side of the breadboard and that it might be a problem with the PDB. Would it be possible to leave the DB9 off the PropStick and use the PDB's RS-232 port instead? (The one next to the RJ-11 jack, not the BS2 programming port.)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

Phil Pilgrim (PhiPi) · 2006-05-04 23:34

The DB9 does hang down below the board. Just leaving the DB9 off would be a problem, because the MAX3232 drives the Propeller's RxD pin in totem-pole fashion. It's not open collector, so you can't parallel it with another signal. To eliminate this, you'd have to omit the MAX3232, but then the reset circuit wouldn't work. One approach would be not to solder the DB9 onto the board, but to connect it with with four wires. This would be pretty easy, since the DB9 has solder cups, and the corresponding pads on the PCB are plenty large. Be sure to use some kind of strain relief near the board, so the wires don't yank on the pads. Plus, if you later change your mind, you can remove the wires, and solder the connector to the board. Once that's done, though, there's no going back.

A better option with less risk of straining the pads would be to solder jumper wires to them and a separate set of jumper wires to the DB9; then use the breadboard to connect the ends.

I can't speak to any other issues surrounding the PDB, since I've never used it. But do watch your voltage levels. The Propeller pins don't like anything above 3.3V without some kind of protection.

-Phil

Sterling · 2006-05-04 23:50

I'm confused. On one post I read the the Propeller can tolerate 5V, and here I read that it can't. I guess it's better to be safe than sorry, but is there a correct answer?

SSteve · 2006-05-05 01:54

My guess is that it can handle 5V on the I/O pins, but not on its input voltage pin. But don't take my word for it--I'm a software guy. We blow things up.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

SSteve · 2006-05-05 01:56

Phil Pilgrim (PhiPi) said...
A better option with less risk of straining the pads would be to solder jumper wires to them and a separate set of jumper wires to the DB9; then use the breadboard to connect the ends.

I like that idea best. Thanks, Phil.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

Phil Pilgrim (PhiPi) · 2006-05-05 02:57

The I/O pins are not 5V-tolerant. Unlike members of the 'LVC logic family, for example, the Propeller has body diodes from each pin to Vdd and Gnd. Inputting a low-impedance 5V signal to any pin risks either a too-high input current or driving Vdd to an unsustainable voltage level, depending on the current-sink capability of the 3.3V regulator. Other threads in this forum describe how input currents can be safely limited using series resistors and other means. One of the more exhaustive discussions can be found here:

http://forums.parallax.com/showthread.php?p=577939

-Phil

Post Edited (Phil Pilgrim (PhiPi)) : 5/5/2006 3:00:18 AM GMT

SSteve · 2006-05-05 03:11

Phil Pilgrim (PhiPi) said...
The I/O pins are not 5V-tolerant.

And that, boys and girls, is why you must never take hardware advice from software guys.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

Paul Baker · 2006-05-05 03:12

Steve, I'm a little curious why you dont breadboard the Propeller sans stick? basically you could get a TO-92 regulator, couple caps, an EEPROM and a crystal and use the RS-232 of the bread-board (using serial resistors).

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
1+1=10

SSteve · 2006-05-05 03:37

Paul: It's mainly because I don't know what I'm doing. I felt like I needed a ready-made solution. I asked about using the Propeller with the PDB in the beginning of April, but I asked about using it with the PDB's USB port. Ken said it wouldn't be practical. I didn't think to ask about using it with the PDB's RS-232 port (and I didn't have the Parallax USB/Serial adapter at the time). But if you don't mind helping me figure out how to do it, maybe I should try that instead of the PropStick. It would certainly save me a lot of soldering. And a lot of troubleshooting my soldering.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

Jon Williams · 2006-05-05 03:38

Per Paul's suggestion, here's how I did that before we made the demo board (see attached).

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

SSteve · 2006-05-05 04:00

Cool. Well I think I'll go that way if I can figure out how.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

Tracy Allen · 2006-05-05 05:02

Phil posted a link to the beta testers forum, which is restricted access. I had posted test results there, which I now copy over to here with a little editing:

Here is data I took on the characteristics of the positive side substrate diode. I didn't test the (-) side diode, but from what Chip said I would expect it to be very similar. The diode starts passing current at 0.4 volts above the 3.3 volt supply, and shows the expected exponential behavior.

attachment.php?attachmentid=41531

In order to keep the current through the substrate diode near 100 microamps with a 5 volts input, you have to use a 10kohm series resistor, and for anything low speed (<1mhz) at the pin, I would opt for that.
With 1kohms the diode has to pass about 1.1 milliamps, which is not bad either for higher speed, up to 10mhz or so.
With 100 ohms the diode would have to pass about 10 milliamps. I pushed it to over 13 milliamps without any evident damage, but I wouldn't push my luck to far! For this kind of speed, a better level translator from 5 to 3.3 volts would be a better option.

It would be wise to include a zener diode or some such clamp to absorb parasitc current, so that it doesn't get into the substrate diode in the first place.

Another consideration is parasitic power. The Propeller is capable of operating at micro power, microamps at 20khz. Most voltage regulators are not capable of sinking current, so it is possible for parasitic power to come into the circuit through a substrate diode and supply power to the Propeller, and even push the power supply voltage up to a level greater than 3.3 volts. You would have to be especially careful about this when using something like a 12 volt input from something like RS232. The Propeller might handle 5 volts, but 12 volts would be a stretch. (disclaimer: I haven't tested that!)

The thread also contained discussion on frequency breakpoints vs the input resistance.

Tracy Allen
www.emesystems.com

Phil Pilgrim (PhiPi) said...
The I/O pins are not 5V-tolerant. Unlike members of the 'LVC logic family, for example, the Propeller has body diodes from each pin to Vdd and Gnd. Inputting a low-impedance 5V signal to any pin risks either a too-high input current or driving Vdd to an unsustainable voltage level, depending on the current-sink capability of the 3.3V regulator. Other threads in this forum describe how input currents can be safely limited using series resistors and other means. One of the more exhaustive discussions can be found here:

http://forums.parallax.com/showthread.php?p=577939

-Phil

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com

Post Edited (Tracy Allen) : 5/5/2006 6:11:14 AM GMT

Phil Pilgrim (PhiPi) · 2006-05-05 05:13

Oh, cripes! Sorry about that. When I did the search to find the thread, I didn't pay attention to which forum it was in. Thanks, Tracy!

-P.

parsko · 2006-05-05 10:38

Gentlemen, Would the attached diagram be valid?

I too, will have the need to deal with this, so I ask. We may want to dedicate a post to this...

-Parsko

Gadgetman · 2006-05-05 10:54

The problem is that the DB9 connector is hanging down below the PCB so that it can't be installed flat on the PDB?

Why not take a 40pin IC-socket(or two) and place them on the PDB first, then the PropStick on top of hem, to raise it up and create the necessary clearance?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Don't visit my new website...

SSteve · 2006-05-12 16:44

My PropSTICK showed up last night. Before putting my non-existant soldering skills to the test, I'm going to try to breadboard the Propeller on my PDB. Between the PropSTICK schematic and Jon's photo earlier in this thread, I can figure most of it out. But there are some things I'm not sure of.

In the PropSTICK schematic the Propeller's A31 is connected to the MAX3232's R2OUT, A30 is connected to T2IN, RSTn is connected through a transistor to R1OUT, and the MAX3232's T1IN is grounded. I want to use the PDB's RS-232 port. It has RTS, TX, CTS, and RX. How do those correspond to R2OUT, T2IN, R1OUT, and T1IN?

Also, in Jon's photo, it doesn't look like the Propeller's Vdd on pin 32 is connected. Is that correct?

On the 24LC256, the PropSTICK has 4.7k resistors between Vdd/SDA and Vdd/SCL. In Jon's photo it looks like he doesn't have a resistor between Vdd and SCL.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

LoopyByteloose · 2006-05-12 17:16

Some of the designs don't have a pullup on both lines. I am not sure why.
It might be the internal pullups alone in the Propeller is faster. It might be economy as the resistors are getting tiny and each one might be a bit of a hassle in production.

The Propeller Demo board shows only the SDA line with a 10K pullup but it is linked to the RESn line.
The Homebrew Board discussion of Mike Cook's talks about the Pro and Con.

It is·here and it will explain a lot.
http://forums.parallax.com/showthread.php?p=579595

I would use both on the PropSTICK until I found a reason not to. You may want to add other chips to the I2C line and need them as the bus gets stacked.
When the day comes that you find a need to remove the SCL pullup [noparse][[/noparse]likely never], it would take very little to bypass it with a bit of wire and a dab of solder.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan

Post Edited (Kramer) : 5/12/2006 6:01:46 PM GMT

SSteve · 2006-05-13 01:39

Thanks, Kramer. I'll go ahead and put in both pullup resistors. But my goal is to not use the PropSTICK or do any soldering. I think I'll start a new thread. I realize now that I've drifted away from my original question.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

SSteve · 2006-05-13 23:31

I broke down and put my soldering skills to the test. Amazingly enough, I managed to assemble a working PropSTICK! It turns out that it works fine on the PDB. It rides high enough that the RS-232 port is above the SX28 socket.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

SSteve · 2006-05-14 00:24

Here's a photo of the PropSTICK happily blinking one of the PDB's LEDs.

attachment.php?attachmentid=41635

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
OS-X: because making Unix user-friendly was easier than debugging Windows

Phil Pilgrim (PhiPi) · 2006-05-14 01:17

Steve,

Thanks for sharing your experience! I've been somewhat at a loss advising people about whether the 'Stick will work with the PDB, since I don't have a PDB myself. Your photo says it all!

Cheers!
Phil

PropStick and Professional Development Board

Comments