AngusMcDoon
Well-Known Member
Contrast working properly, controlled by a TL061 op amp.
cheaper large instrument displays - mast mount
- Thread starter ShaunG
- Start date
Tom_Brewis
Active Member
Thats brilliant. Does this get rid of the 10k pot or do you set it at a sensible level?
What is the RS part number, is it
http://uk.rs-online.com/web/p/operational-amplifiers/6611300/
Or
http://uk.rs-online.com/web/p/operational-amplifiers/7092019/
Or
http://uk.rs-online.com/web/p/operational-amplifiers/6611306/
And do i need anything else?
The RF modules arrived yesterday, family permitting I shall get the soldering iron out!!!
What is the RS part number, is it
http://uk.rs-online.com/web/p/operational-amplifiers/6611300/
Or
http://uk.rs-online.com/web/p/operational-amplifiers/7092019/
Or
http://uk.rs-online.com/web/p/operational-amplifiers/6611306/
And do i need anything else?
The RF modules arrived yesterday, family permitting I shall get the soldering iron out!!!
AngusMcDoon
Well-Known Member
355-5042 cheaper as you don't have to buy 10. Made by ST rather than TI, so the Feng Shui will be better given that it's a ST processor driving it
Get a few resistors every 10k from 10k to 150k, costs buttons and will be useful. You only need small wattage ones. Your display is different from mine and you may need different value resistors as you voltages will be slightly different.
10k pot used one-off to calibrate the mid range of contrast adjustment on the slider with mid-range between all white and all black on the screen. Then software control after that.
If the displays are all similar it may be possible to do away with the pot and use a fixed resistor, but I don't know how repeatable they are. A small 10k trim pot on the board would not be a problem though - calibrate at first power up, then close the box and forget.
Tom_Brewis
Active Member
That sounds good. I will get the op amp when the shops open. Still wiring up the remote.
AngusMcDoon
Well-Known Member
The remote handset now runs from batteries...
I have reduced the power consumption down to 45 microAmps in standby. From 2 AAA Duracells this will give a life of 3 years, or 18 months from a lithium coin cell. The 18LF24K22 is lower power but otherwise the same,so I might try that instead. It costs another 17p.
This is now using the cheaper 18F24K22 processor. The components are (inc VAT, buying 10) ...
PIC18F24K22 £1.83 (or £2.00 for low power version)
RF module £1.12
Capacitors/resistors 6p
LED 5p
AAA battery holder 41p
These are for through-hole components. Surface mount would be cheaper. A PCB would cost about £1.50 if 10 were made by, for example, iteadstudio. Cost drops considerably for larger numbers.
To the main display I have added a feature to show the raw data coming in. This shows the NMEA sentences and the Seatalk packets in hex...
The rolling display can be stopped and started. Seatalk is consistent, but NMEA is not. If someone reports that their message is not supported then using this feature it can be seen what is actually coming in and makes adding support for strange sentences easier.
Shaun seems to have gone quiet. What's happening with the boxes? The project is no good without the plastics. I feel like I'm sailing single handed here!
I have reduced the power consumption down to 45 microAmps in standby. From 2 AAA Duracells this will give a life of 3 years, or 18 months from a lithium coin cell. The 18LF24K22 is lower power but otherwise the same,so I might try that instead. It costs another 17p.
This is now using the cheaper 18F24K22 processor. The components are (inc VAT, buying 10) ...
PIC18F24K22 £1.83 (or £2.00 for low power version)
RF module £1.12
Capacitors/resistors 6p
LED 5p
AAA battery holder 41p
These are for through-hole components. Surface mount would be cheaper. A PCB would cost about £1.50 if 10 were made by, for example, iteadstudio. Cost drops considerably for larger numbers.
To the main display I have added a feature to show the raw data coming in. This shows the NMEA sentences and the Seatalk packets in hex...
The rolling display can be stopped and started. Seatalk is consistent, but NMEA is not. If someone reports that their message is not supported then using this feature it can be seen what is actually coming in and makes adding support for strange sentences easier.
Shaun seems to have gone quiet. What's happening with the boxes? The project is no good without the plastics. I feel like I'm sailing single handed here!
Last edited:
ShaunG
Well-Known Member
Sorry for the quietness..I have been unwell the last couple of weeks and missed my meeting with the plastics people..I will be on to it first thing in the new year
Tom_Brewis
Active Member
Last edited:
Tom_Brewis
Active Member
RF Module working
After quite a lot of mistakes I now have the RF module working with the YAPP address software.
At one point I was trying to create a stand alone power supply for the remote., unfortunatley whilst investigating the outputs from the PICKIT 2 I managed to create a short and all stopped working. I thought i had blown the RF module but actually i had corrupted the code in the PIC device. A reflash and all was well.
Angus when I try to go battery on the remote, where do i wire the PICKit2 pin 1?
Thanks
All the best
Tom
Hope your feeling better Shuan.
After quite a lot of mistakes I now have the RF module working with the YAPP address software.
At one point I was trying to create a stand alone power supply for the remote., unfortunatley whilst investigating the outputs from the PICKIT 2 I managed to create a short and all stopped working. I thought i had blown the RF module but actually i had corrupted the code in the PIC device. A reflash and all was well.
Angus when I try to go battery on the remote, where do i wire the PICKit2 pin 1?
Thanks
All the best
Tom
Hope your feeling better Shuan.
AngusMcDoon
Well-Known Member
The PICKit2 is tolerant of shorts - fortunately!
It's quite difficult for me to post at the moment because I'm in Culkein-Drumbeg at the top of Scotland where the only internet is outside at the top of the nearest hill, and it's blowing 7 bells right now. My netbook keeps on blowing off my lap!
For battery power no need for any connection between pins 1 on the processor and 1 on the PICKit. Leave pin 1 on the processor disconnected. What I have found though is that you need a 10uF capacitor across the power lines where the battery wires go into the board. You can see the small black cylinder on my photo above. When powering up by the PICKit the power comes on nice and cleanly. Powering up by pushing a battery in does not, and the processor/RF module can get in a funny state. The capacitor smooths this out and so far it has powered up first time every time.
I have now reduced the power consumption to 13 microAmps which will give a battery life of 6 years for the coin cell, 11 years for 2xAAA. I don't think it is necessary to go to the low power processor, which if the datasheet is to be believed would give a battery life of over 5000 years!
I will update the schematic and send new software when I'm back in civilization. Tom I take it you are using the 18F26K22? I am using the 18F24K22 now. The only difference is less flash space for less cost. I can do a 26K22 version as well as a 24, it's only a few lines of configuration code different. Microchip do a 18F23K22 for less money, but RS and Farnell don't sell it.
Shaun hope you are back to normal soon. You have some catching uop to do Buttons on the display are so last year!
Fingers now frozen so back inside.
It's quite difficult for me to post at the moment because I'm in Culkein-Drumbeg at the top of Scotland where the only internet is outside at the top of the nearest hill, and it's blowing 7 bells right now. My netbook keeps on blowing off my lap!
For battery power no need for any connection between pins 1 on the processor and 1 on the PICKit. Leave pin 1 on the processor disconnected. What I have found though is that you need a 10uF capacitor across the power lines where the battery wires go into the board. You can see the small black cylinder on my photo above. When powering up by the PICKit the power comes on nice and cleanly. Powering up by pushing a battery in does not, and the processor/RF module can get in a funny state. The capacitor smooths this out and so far it has powered up first time every time.
I have now reduced the power consumption to 13 microAmps which will give a battery life of 6 years for the coin cell, 11 years for 2xAAA. I don't think it is necessary to go to the low power processor, which if the datasheet is to be believed would give a battery life of over 5000 years!
I will update the schematic and send new software when I'm back in civilization. Tom I take it you are using the 18F26K22? I am using the 18F24K22 now. The only difference is less flash space for less cost. I can do a 26K22 version as well as a 24, it's only a few lines of configuration code different. Microchip do a 18F23K22 for less money, but RS and Farnell don't sell it.
Shaun hope you are back to normal soon. You have some catching uop to do
Buttons on the display are so last year!Fingers now frozen so back inside.
Tom_Brewis
Active Member
Yes I am using 18F26K22.
I tried removing pin1 connection and although I had power on he board it would not work. I will fit a 10uF cap across the battery.
So I can take it all to the boat to test, on the main display can I just feed 12v through a 5v regulator to 5v on the ST discovery board?
Do wait till the weather abates before replying.
I tried removing pin1 connection and although I had power on he board it would not work. I will fit a 10uF cap across the battery.
So I can take it all to the boat to test, on the main display can I just feed 12v through a 5v regulator to 5v on the ST discovery board?
Do wait till the weather abates before replying.
Tom_Brewis
Active Member
Remote now running on battery power. 2 AAA batteries.
AngusMcDoon
Well-Known Member
The low power software version should run on a 3V lithium coin cell for an acceptably long period, but at this stage they are more difficult to arrange holders for.
Tom_Brewis
Active Member
Contrast Control
Hi Angus et al
Happy new year
I have received the op amp. I cant work out the wiring, the 10k pot is just wired through the display board, the op amp presumably needs some control from the discovery board. Please can you tell me the pin outs.
I have got an abs box and hope to take it all to the boat at the week end to see if it all works with real data.
Can I feed 12v through a 5v regulator to the discovery board to the 5v pin, to power it up instead of USB?
To feed seatalk in do i just make the Thomas Knauf circuit?
Thanks
Tom
Hi Angus et al
Happy new year
I have received the op amp. I cant work out the wiring, the 10k pot is just wired through the display board, the op amp presumably needs some control from the discovery board. Please can you tell me the pin outs.
I have got an abs box and hope to take it all to the boat at the week end to see if it all works with real data.
Can I feed 12v through a 5v regulator to the discovery board to the 5v pin, to power it up instead of USB?
To feed seatalk in do i just make the Thomas Knauf circuit?
Thanks
Tom
AngusMcDoon
Well-Known Member
There's a bit of voltage biasing that needs doing to connect the op-amp. I'll draw it tomorrow.
I need to look at the Disco board datasheet to see if the 5V pins are connected directly to the USB power line. I expect they are.
I used Thomas Knauf's wiring and it works fine for me.
I haven't done any Seatalk tests recently - been doing more with NMEA. There's a commented out line in init.c (call to init_mast_display()) that needs to go back in for Seatalk to work.
I need to look at the Disco board datasheet to see if the 5V pins are connected directly to the USB power line. I expect they are.
I used Thomas Knauf's wiring and it works fine for me.
I haven't done any Seatalk tests recently - been doing more with NMEA. There's a commented out line in init.c (call to init_mast_display()) that needs to go back in for Seatalk to work.
Tom_Brewis
Active Member
Thanks. I tried feeding 12v in via a 5v regulator and it all works fine.
I will find the comented out bit and un comment.
Will build seatalk circuit.
Looking forward to the drawing for the op amp
Thanks
I will find the comented out bit and un comment.
Will build seatalk circuit.
Looking forward to the drawing for the op amp
Thanks
AngusMcDoon
Well-Known Member
This is the power circuit on the Disco board.
U5V_ST_LINK is the USB line.
EXT_5V is the edge pin marked 5V.
The diode is to prevent supplying current to the USB power line if EXT_5V > U5V_ST_LINK .
So powering by applying 5V to the 5V pin is fine.
Tom_Brewis
Active Member
Thanks. That's why I didn't blow it up. Grin
Jabamusic
Well-Known Member
Fantastic developments...but how far along are we (you!)
I continue to follow this thread with fascination, interest and hope as I suspect do many others........
But I was wondering (especially as the last months worth of posts have been incredibly technical) where things are up to?
Is there any chance somebody could give us a sit rep (without the detail) on what they believe has been thus far achieved and then perhaps what is still required? You never know we might be able to help out (fat chance probably - but I like to believe I can be useful sometimes !!).
No pressure guys, just me being keen !!
I continue to follow this thread with fascination, interest and hope as I suspect do many others........
But I was wondering (especially as the last months worth of posts have been incredibly technical) where things are up to?
Is there any chance somebody could give us a sit rep (without the detail) on what they believe has been thus far achieved and then perhaps what is still required? You never know we might be able to help out (fat chance probably - but I like to believe I can be useful sometimes !!).
No pressure guys, just me being keen !!
Pinnacle
Well-Known Member
AngusMcDoon
Well-Known Member
Display software all working for NMEA for the messages chosen to support. Easy to add support for more messages if required now that the framework is in place. About 30 NMEA sentences are supported.
Seatalk support was working for about half the messages last time Seatalk was tried by me. The Seatalk message handler has been extensively rewritten and just needs testing as there may be a minor bug or two. Now supports pretty much the same as NMEA.
Power optimization for display software not started. I will do this if any devices end up on boats, as it doesn't add anything to functionality but is quite tricky to do.
Remote handset software complete and working, running on a PIC 18F26K22 or cheaper 18F24K22. Optimized for ultra-low power running giving a coin cell predicted lifetime of 3 years.
Reception of remote control messages on the display complete and working.
All graphics routines, persistent settings storage, setup and general user interface complete.
Hardware design for the remote handset complete, including schematic but excluding PCB layout. This is using a pre-assembled RF module and aerial costing about £2 for a pair.
Hardware design for display is still using the off-the-shelf Discover board, plus a few extra support components. Problem of software control of contrast resolved. NMEA input is still RS-232 not RS-422, but that is just a chip change. Schematic diagram for the display done in parts. Creating a full hardware design involves copying the Discovery board as this is an open reference design, but cannot be done without a multi-layer PCB being designed.
The major component where I don't know what is happening is the casing. Shaun hasn't posted recently.
There are 4 working displays (me 2, Tom 1, Shaun 1) and 2 working remote handsets (me 1, Tom 1). Only 2 Discovery boards were harmed in the process
So, in summary, software almost complete, hardware design (excluding PCB) almost complete, PCB layout not started, case Shaun to report.
This is of course only to hobby standards. Nothing done on EMC compliance or European standards testing, although best practice techniques have been used in software for EMC.
Seatalk support was working for about half the messages last time Seatalk was tried by me. The Seatalk message handler has been extensively rewritten and just needs testing as there may be a minor bug or two. Now supports pretty much the same as NMEA.
Power optimization for display software not started. I will do this if any devices end up on boats, as it doesn't add anything to functionality but is quite tricky to do.
Remote handset software complete and working, running on a PIC 18F26K22 or cheaper 18F24K22. Optimized for ultra-low power running giving a coin cell predicted lifetime of 3 years.
Reception of remote control messages on the display complete and working.
All graphics routines, persistent settings storage, setup and general user interface complete.
Hardware design for the remote handset complete, including schematic but excluding PCB layout. This is using a pre-assembled RF module and aerial costing about £2 for a pair.
Hardware design for display is still using the off-the-shelf Discover board, plus a few extra support components. Problem of software control of contrast resolved. NMEA input is still RS-232 not RS-422, but that is just a chip change. Schematic diagram for the display done in parts. Creating a full hardware design involves copying the Discovery board as this is an open reference design, but cannot be done without a multi-layer PCB being designed.
The major component where I don't know what is happening is the casing. Shaun hasn't posted recently.
There are 4 working displays (me 2, Tom 1, Shaun 1) and 2 working remote handsets (me 1, Tom 1). Only 2 Discovery boards were harmed in the process
So, in summary, software almost complete, hardware design (excluding PCB) almost complete, PCB layout not started, case Shaun to report.
This is of course only to hobby standards. Nothing done on EMC compliance or European standards testing, although best practice techniques have been used in software for EMC.
Last edited:
