Searching \ for '[PIC] Circuit Comment and Critique' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'Circuit Comment and Critique'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] Circuit Comment and Critique'
2007\01\24@104009 by Rolf

face picon face
part 0 44 bytes
his is a multi-part message in MIME format.
part 1 5480 bytes content-type:text/plain; charset=ISO-8859-1; format=flowed (decoded 7bit)

Hi all.

I am looking for some comments and criticisms on the design of a 418MHz
wireless transmitter for a hobby project.

I am nearing the end of my first "Major" project. Consists primarily of
an "advanced" shutter control system for my camera. Will control long
exposure photos, time-lapse, and other modes. One 'mode' is to have a RF
wireless remote for the device. The camera has an IR remote already, but
this requires the operator to be in front of the camera, and is limited
range (about 5 meters - with the OEM remote - my home-built IR remote
has about 30m range).

The wireless component is new to me in many ways. I decided to go
'medium simple', and use the chips from LINX tech: the TXM-418-LR and
the RXM-418-LR as 418MHz transmitter and receiver respectively, using
the LINX "splatch" antenna. I have carefully implemented everything as
much as possible according to the datasheets and application notes from
LINX. The microstrip trace on my proposed PCB connecting th TX and
Antenna will be less than 0.2", there is as large a ground-plane as
possible, etc.

LINX 'advertise' that if you have a 1-wire 'serial' communication link
between to processors, you can simply replace the wired link with the TX
and RX modules, and it will work up to 10kbit. Well, I have the
prototype system running with a single wire (in a breadboard), and it
works, so I want to now implement it with RF instead of the wire. My
plan is to build the 'final' version of the remote control (TX), and a
"break-out board" receiver module to help me debug the RX side on a
breadboard.

The attached circuit diagram is of the transmitter, as well as the top
and bottom board layers. The entire board is 1.9" x 1.2", and fits in
the C-4 box from Serpac. All produced with Eagle 4.16

The Major components are:
PIC10F206
TXM-418-LR TX chip
the matching LINX "Splatch" SMT antenna
A Momentary switch N.O., pulled low with a weak pull-down.
a CR2032 Li. Cell.
a Schottky diode (will be explained later)
a 5-pin In-Circuit PIC programming header (The pin order is non-standard
- it is easy to route on the PCB though).

I have implemented a concept of an "Analog Ground". All the ground
connections for both the TX and the antenna are fed through to the drain
of an NMOSFET. The ground-plane of the entire board will be this
"Analogue Ground", and NOT the negative terminal of the battery, or Vss
of the PIC.

The basic operation is as follows:
The PIC10F206 sleeps in normal operation. This consumes 0.1uA typically.
MCLR is pulled high with internal weak pull-up. GP0, GP1, GP2 is set as
high-impedance input and will be pulled low.
Normally the NMOSFET Gate is pulled low by the 1M to ground. This
"disconnects" the "Analog Ground" from GND, and isolates the transmitter
and antenna from the rest of the circuit.

Pressing the button wakes the PIC, which sets GP1 and GP2 to output. It
sets GP1 high, saturating the FET, and connecting the TX and antenna to
ground, thus powering up the TX (and illuminating the LED).
After a delay to allow the TX to stabilize, the PIC bit-bangs a serial
data stream over GP1 to the TX (2 bytes + sync at 1200bps is less than
20ms).
After the transmission, it reverts back to it's high impedance sleep
mode (effectively powering off the TX, and disconnecting it's ground).

There are three major considerations I have when I designed the circuit
this way:
The TX is Vmax of 3.6V.
I will almost certainly need to re-flash the PIC a number of times as I
work out issues (in fact, the first program of the chip will be in
circuit - SOT 23-6).
Power consumption must be low.

In order to keep low power consumption I use the MOSFET to low-side
switch the entire TX system. Only leakage current will happen.
Additionally, all inputs are put in high-impedance, and pulled to ground.
MCLR is pulled high (internal weak pull-up).

The ICSP header will be available for (re-)programming easily. The
Schottky diode protects the 3V cell from the 5V programming voltage.

My real concern is the 3.6V Max for the TX chip. When programming I have
2 concerns: ICD2 will pull Vdd to 5V. Because the ground part of the TX
circuit is isolated by the MOSFET, I believe the TX will be isolated
from the 5V. There are three possible problems with this though:
First, will GP2 be set output-high at any point by programming the chip?
Second, After programming, the PIC will be reset (remeber, Reset is what
happens after a sleep in the PIC10F series - it does not continue from
the instruction after SLEEP), before starting the transmissin process,
the PIC checks to ensure that the trigger pin GP0 is high, but perhaps
the ICD2 will leave the data line high (or will maybe pulse it or
something).
Third, 5V programming Vdd will charge the capacitor... activating the
circuit after programming will cause the capacitor's charge (at 5V) to
potentially destroy the TX.

Is there a better way to ensure that programming the PIC does not damage
the TX?

Thanks for any comments/critiques of the board.

I know this is a big request to put on the list, and I appreciate that
this is a long post... but, the next step is to build a 3" x 4.5" board
which will be much more densely populated, and I need to get the process
right before I mess something up there.

Thanks all

Rolf


part 2 11408 bytes content-type:image/png; (decode)


part 3 10674 bytes content-type:image/png; (decode)


part 4 11020 bytes content-type:image/png; (decode)


part 5 35 bytes content-type:text/plain; charset="us-ascii" (decode)

2007\01\24@135842 by Vasile Surducan

face picon face
For free ?

:)
Vasile

On 1/24/07, Rolf <spam_OUTlearrTakeThisOuTspamrogers.com> wrote:
{Quote hidden}

> -

2007\01\24@141739 by Richard Prosser

picon face
Just on a quick review.
It might pay to add a bit more decoupling as close to the 2 ICs as possible.
As far as the programming voltage is concerned - could you connect the
TX chip to the battery via a separate diode. This would prevent it
being powered by more than the battery voltage less a diode drop.
There may be an opportunity for 5V to be applied to an input however.
e.g. the data Input?. This may need protection using a series R and a
diode or zener.

RP

On 25/01/07, Rolf <.....learrKILLspamspam@spam@rogers.com> wrote:
{Quote hidden}

> -

2007\01\24@142708 by Maarten Hofman

face picon face
Redwood Shores, 24 januari 2006.

Dear Rolf,

All documentation I have seen so far states that the PIC always sets the I/O
pins to input, preferably analog, if available. So the only time that an I/O
pin would be output and high is if your own code does this. It is possible
that your code will run during programming (I've seen it happen with my own
code) so just make sure that it doesn't make the output high at that time.

Alternatively you could protect various parts using a zener diode, although
I have little experience with that.

Greetings,
Maarten Hofman.

2007\01\24@144440 by Rolf

face picon face
Well, uhm, as I said, I know it was a long post, with lots of detail,
and so on. I don't expect it to be realistic for everyone to dive in and
grasp the whole project, etc. I tried to be up-front about that.

I was hoping that someone more experienced than me (like 99% of the
people on the list) would be able to spend the time to go through it.
And I mean the "hoping" part, I am sure that, for most people here, the
time required would be excessive.

So, in short, yes, for free... ;-) My options are to beg for an
experienced set of eyes, or to build it myself, and risk failure, and
'earn' the experience the hard way. Failure would create a substantial
dent in my hobby budget just in the parts alone (about $50 parts in the
transmitter...).

I have done a lot of homework to get where I am now, and, while I have
come some ways, the finer details are learned expensively through
experience, or listening (rather cheaper) to experienced people.... I am
hoping to learn through the latter, for once.

I hope that my original post did not imply I expect everyone to put my
request at the top of a priority list...

Rolf

Vasile Surducan wrote:
{Quote hidden}

2007\01\24@150123 by Rolf

face picon face
part 0 44 bytes
his is a multi-part message in MIME format.
part 1 1266 bytes content-type:text/plain; charset=ISO-8859-1; format=flowed (decoded 7bit)

Shoot, that's so logical.

I can put the TX on it's own diode from the battery, and I can use a
MOSFET to control the Data Line. This will provide 3.6V safety at all
times. I will have to "reverse" the sense of the data transmission, 1
now means 0.

Thanks, that makes a big difference. I can put a second Cap for the TX
in as well, It will have to be based on the "real" ground, not the
analog ground. I will ensure that the caps are placed near the IC's, I
though I had placed it near the PIC, but I hadn't.

Thanks again. See, updated schematic attached.... pretty picture. It's
more obvious now seeing the isolation through the two MOSFETS.

Rolf

Richard Prosser wrote:
{Quote hidden}


part 2 31136 bytes content-type:image/png; (decode)


part 3 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)

2007\01\24@151101 by Bob Blick

face picon face
Hi Rolf,

You are switching the ground for your RF circuit and antenna. Even if you
added additional AC coupling to ground it still seems like a bad idea to
me and something I would never do unless circumstances forced me, and then
I'd test it both ways.

Cheerful regards,

Bob


2007\01\24@151158 by Anno, Jeff

flavicon
face
Rolf,

I think that asking for opinions is fine. You didn't sound like you
were demanding that others do your homework for you.  You obviously
spent some time and just want another set of eyes on it.  Where else
can you get such a great resource than the piclist?

I think that the "For free?" reply was just in good fun (it was followed by
a smiley face).

If I wasn't in the same situation as you I would offer some advice... But
I think that people who are more advanced than a hobbyist (which is all
that I am) can probably add a more helpful critique.

One helpful tip... to increase your hobby budget skip lunch!  ;)

-Jeff




{Original Message removed}

2007\01\24@180230 by Richard Prosser

picon face
I don't think it will matter in your case, but don't forget that
adding the MOSFET to isolate the data line will invert the signal.

RP

On 25/01/07, Rolf <.....learrKILLspamspam.....rogers.com> wrote:
{Quote hidden}

> -

2007\01\25@060844 by Michael Rigby-Jones

picon face


>-----Original Message-----
>From: EraseMEpiclist-bouncesspam_OUTspamTakeThisOuTmit.edu [piclist-bouncesspamspam_OUTmit.edu]
>On Behalf Of Bob Blick
>Sent: 24 January 2007 20:11
>To: Microcontroller discussion list - Public.
>Subject: Re: [PIC] Circuit Comment and Critique
>
>
>Hi Rolf,
>
>You are switching the ground for your RF circuit and antenna.
>Even if you added additional AC coupling to ground it still
>seems like a bad idea to me and something I would never do
>unless circumstances forced me, and then I'd test it both ways.
>

That was one of the first things that I noticed and I agree it's asking for problems.  For the cost of an extra transistor, a P channel MOSFET or even a cheap PNP bipolar switching the positive line with some decoupling after it would be a much more comfortable solution.

Regards

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2007\01\25@064056 by Peter P.

picon face
I think that switching the ground on the RF part as shown is not a problem if
you add a decoupling cap directly on the RF chip. This should work fine as lng
as the antenna is reasonably well matched to the transmitter and the unit ground
is actually floating (handheld or fully insulated). Adding a second, smaller
decoupling cap across the FET DS can also help to avoid problems.

Peter P.


2007\01\25@081924 by Rolf

face picon face
Hi all, again.

Thanks Mike, Peter, Bob for the comments on the ground-plane switching.

I have revised the circuit, and removed the ground-switching MOSFET
entirely. I had put it in there to isolate the RF components when at 5V,
but, with the diode idea from Richard, I really don't need to worry
about that.

As for the drain current of the TX, I can leave it connected to the 3V
cell as long as I power it down by pulling the PDN pin low. It has a 5nA
Ipdn typically. Instead of having the FET control the ground plane, I
now will control the PDN pin, with a P-MOSFET pulling it up to Vcc when
required (and a weak pull-down otherwise).

With that change, the two FETS  and the diode on the cell isolate the
PIC side of the circuit completely from the RF side. Programming the PIC
can not possibly  damage the TX components.

In addition, I am no longer switching the ground-plane, and both RF and
PIC components share a common ground.

Fortunately, the RX side of the RF link is 5V tolerant.... ;-)

Thanks again

Rolf


Michael Rigby-Jones wrote:
>  
>> {Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...