• 10 dec 2017: forum version update. In case of issues use this topic.
  • 30 nov 2017: pilight moved servers. In case of issues use this topic.
Hello There, Guest! Login Register


Low-profile PCB design
#11
pilight never sends and receives at the same time, so connecting both to the antenna should work. A pcb antenna is just a lane of 17cm across the pcb.
 
Reply
#12
When we are (i hope seriously) creating the PCB, i have some feature requests. I would like to add (when possible):
1. A IR receiver (preferable a TSOP4838).
3. A voltage regulator.
4. A jumper that allows us to connect either the Superheterodyne receiver.
[Image: $T2eC16VHJGYFFlLe3qSvBSQ9hhRJo!~~60_12.JPG]
Or the cheap one:
[Image: $_35.JPG]
(We had good experiences with the last one when using a voltage regulator).
5. A RJ11 port and a SN74HC14N.

Please notice that the sender and the receiver should be connected by pushing them into a header because they are too sensitive to properly solder.

Can you tell me if that's (all) possible.
 
Reply
#13
(12-08-2014, 09:44 PM)curlymo Wrote: When we are (i hope seriously) creating the PCB, i have some feature requests. I would like to add (when possible):
1. A IR receiver (preferable a TSOP4838).
2. A voltage regulator.
3. A jumper that allows us to connect either the Superheterodyne receiver. (img) Or the cheap one: (img) (We had good experiences with the last one when using a voltage regulator).
4. A RJ11 port and a SN74HC14N.

Please notice that the sender and the receiver should be connected by pushing them into a header because they are too sensitive to properly solder.

Can you tell me if that's (all) possible.

I've not been working on the PCB much the last few days (the SMA-plug is gone and it's replaced by two PCB antennas in my latest draft though). In response to your questions:
1. Putting a TSOP48XX on the board is probably no problem.
2. What would you want to use the regulator for?
3. I took a quick look at the pinouts of both receivers and I'm pretty sure we can do that with a (few) jumper(s). We could easily silkscreen the correct jumper settings for different types on the board.
4. What would you want to do with the RJ11 and SN74HC14N?
 
Reply
#14
The regulator is meant to reduce noise from the receiver. The RPi 5v header is pretty unregulated.
The RJ11 and the SN74HC14N is meant for future smart energy meter reading.
 
Reply
#15
This is the schematic of the voltage regulator by our user John1988. Feedback is of course welcome:
[Image: voltageregulator.jpg]
 
Reply
#16
Here you have the smart meter schematic as well:
[Image: smartmeter.png]
 
Reply
#17
(12-15-2014, 05:49 PM)curlymo Wrote: Here you have the smart meter schematic as well:
--image--

Thanks for the schematics, I don't think including these would be an issue. However, I may want to exchange the sn74hc14n for another logic inverter, as this is quite a big IC to put on the board, especially since we're only using a single inverter port. Is there any specific reason to prefer the sn74hc14n?
I did some more checking on the ZR-3A, and it seems that if we're not connecting the pin labeled DER on the RXB6 (and I don't know what it's for, so that's the case), the ZR-3A should actually be pin compatible. Furthermore, I've redesigned the board using EAGLE (Fritzing was getting on my nerves) and I've removed the PCB antenna from my design. I'm not sure I can design a good PCB antenna with the current board size and not risk some serious interference. I've replaced the antennas with through holes to mount a wire antenna on. I'll be posting my new design somewhere in the coming days.
Looking at this got me wondering, however. The datasheet for the jmr-tx-1 says that it's absolute maximum rating is 5V, so when using a 5.2V power supply, we might damage it. Shouldn't we be powering this from the 3.3V header?
By the way, the logic levels of the ATTiny being powered at 5V would be around 4.3V, which is more than the 3.3+5% the RPi is supposed to be tolerant of, am I right in guessing the 1K resistors are there for this reason?
 
Reply
#18
(12-16-2014, 09:11 AM)martenjacobs Wrote:
(12-15-2014, 05:49 PM)curlymo Wrote: Here you have the smart meter schematic as well:
--image--

Thanks for the schematics, I don't think including these would be an issue. However, I may want to exchange the sn74hc14n for another logic inverter, as this is quite a big IC to put on the board, especially since we're only using a single inverter port. Is there any specific reason to prefer the sn74hc14n?
It just has to be a hex inverter Smile

Quote:I did some more checking on the ZR-3A, and it seems that if we're not connecting the pin labeled DER on the RXB6 (and I don't know what it's for, so that's the case), the ZR-3A should actually be pin compatible.
That would be great!

Quote:Furthermore, I've redesigned the board using EAGLE (Fritzing was getting on my nerves) and I've removed the PCB antenna from my design. I'm not sure I can design a good PCB antenna with the current board size and not risk some serious interference. I've replaced the antennas with through holes to mount a wire antenna on. I'll be posting my new design somewhere in the coming days.
I'm curious of course.

Looking at this got me wondering, however. The datasheet for the jmr-tx-1 says that it's absolute maximum rating is 5V, so when using a 5.2V power supply, we might damage it. Shouldn't we be powering this from the 3.3V header?
[/quote]
The voltage regulator should already drop the voltage a little bit i think.

Quote:By the way, the logic levels of the ATTiny being powered at 5V would be around 4.3V, which is more than the 3.3+5% the RPi is supposed to be tolerant of, am I right in guessing the 1K resistors are there for this reason?
I don't use the resistors but saw some blog entries that state that the 1K resistors should always be safe w/o effecting operations.
 
Reply
#19
I've done some more work:
  • I did some looking around and can't find any smaller Schmitt-trigger inverter that's not SMD, so I've used the sn74hc14n.
  • I've placed the interface for the RJ11 connector in a screw terminal block like this one for two reasons: first, space-wise it will be difficult placing a full connector on the board without making the board too big to be compatible with both the model B and the B+; and secondly, the PCB is meant to fit inside a case. When adding a connector to it, you'll have to cut a hole in the case to accommodate it. With a terminal block you could simply connect the wires from the cable directly to the board, or if you really want a connector, buy a panel mount version and connect that to the terminal block.
  • I've added a trace from the receiver directly to a header on the RPi (with a 1K resistor in between). I'm hoping this will enable us to bypass the BPF when using pilight-debug, or would this not work?
  • I added a nice pilight logo to the silkscreen
  • I checked with mijnprintplaat.nl, and when ordering 10 pcs, their price is about €79,00 incl. VAT
 
Reply
#20
In regard to your points:
1. Ok.
2. The whole point of of course using a RJ11 socket is the ease of it.
3. Don't.
4. Smile
5. Does mijnprintplaat.nl also pre-solder the common parts? That would be sort of an requirement.

Looking at your circuit:
1. The IC2A 2 output should go to the RxD. It doesn't seem like it does. The RxD is at the same side as the 5v.
2. Is there a reason to use different default sender / receiver pins?

Some other questions:
1. Can the IC1 actually become a 8pin IC socket?
2. Are the sender / receiver connected headers?
 
Reply
  


Forum Jump:


Browsing: 2 Guest(s)