I suspect "mitre" means "to remove or to round off". (I couldn't fint it in the dictionary)
If this is the case, I will have to redraw it with Eagle because my PCB design software is very basic and doesn't allow someone to round off the corners.
Yes, it means putting 45 degree bends in the tracks. Try "miter", that's the American spelling. It won't matter much with your board, but it makes it look more professional, and can prevent etching problems.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
At first I couldn't discern the red and green trace colors; I've done lots of boards (maybe I'm too color-blind) and use high contrast colors like red and blue, I couldn't tell you had a two sided board at first, so found lots of shorts. My bad eyes. Thread title threw me off; "Homemade singe side..."
There is still a 90° bent at upper-right. And an unconnected trace from the top pin of the regulator to the power jack.
Do you have any program to verify the layout connections vs. the schematic? Invaluable to find unconnected pins, shorted traces, traces too close, etc.
Thank you for examining the design so carefully. I have made the rectification.
No, my program doesn't make the verification with the schematic. Eagle does, but I don't like using Eagle because I find it so difficult to discover the right component within a huge library and I have never learnt to build my own library. Probably a bit too lazy !
You may want to think about adding a .1uf or .01uf cap across the supply lines at the EEPROM and at the Propeller chip. I've gotten into the habit of doing that for all IC's and it is pretty common practice. Doesn't add much to the cost.
There are two more 90°s; below pins 9 and 12 of the Prop. The ground one could be helped with a via and the trace from the left end of the big cap on the other side. I've done enough 4-layer board, power/ground on the inside and signal traces on outer sides for access visually and for mods if needed, so many of this sort of effect was absent.
I did one large 4-layer pcb, about 15 x 18", that would have been horrible without any open/short and proper continuity checking available, as it went through numerous iterations while being laid out (the designer wasn't finished yet the company head wanted the layout started way before the majority of the design placement was thought out by him. some 'fun', NOT!).
Harley said...
There are two more 90°s; below pins 9 and 12 of the Prop. The ground one could be helped with a via and the trace from the left end of the big cap on the other side.
The board is single side. The red lines are the copper tracks underneath. The green lines are wires on the top. The kaki lines are the components or the edge of the board. I am sorry but I cannot change the colours.
So, I don't see what I can do to the 9O° angles below pin 9 and 12 of the Prop.
The two 0.1uF caps you have added look to be very large. Either theu are can electrolytics or far too high a voltage rating. I use ceramic disk or surface mount ones (if you don't mind it being non-through hole). The disk ones are usually small enough to hide on the gaps of the IC base and the surface mount ones just fit across the tracks and do not present any thickness to the track side.
Many people, including me, try to get a additional 47-100uF Tantalum cap as close to the supply pins as possible, again there is usually space in the IC base.
I have fallen in love with the copper area pour, for a ground plain mostly. It has to help with shielding, and uses less Ferric Chloride too.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Why did I think a new, more challenging, job was a good idea ??
Post Edited (Toby Seckshund) : 8/3/2010 9:17:27 AM GMT
The board is single side. The red lines are the copper tracks underneath. The green lines are wires on the top. The kaki lines are the components or the edge of the board. I am sorry but I cannot change the colours.
So, I don't see what I can do to the 9O° angles below pin 9 and 12 of the Prop.
Ooops! I had missed that point of the green color for the wires on pcb top.
Just ignore my comment about 90° trace bends. (As a friend used to say many times "First mistake".)
You might be interested in an old Propeller system layout I did. It fits all the routing on 1 layer with no jumpers. (see attached graphic) The eeprom, programming connector, and regulators are at the top and bottom of the propeller chip with full access to all I/O pins from the side. I'd make two changes now when I used this. First I'd connect the brown out detector with a cut-jumper in case I wanted to run at ultra low power. Second I'd find space for a pull up resistor on the reset line.
Lawson
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Lunch cures all problems! have you had lunch?
Lawson, that's a good layout. You could add some connectors to pins for the ground, power, and I/O pins and have a decent bare bones prop for bread boarding.
It is a good idea to have a jumper on the BOEN pin. I should add one.
Also I will have to check what is needed to use the RS232 for communication with an other controller, terminal or PC (other than programming purpose). Perhaps it would be worth forseeing one or two more jumpers so that I can use the same connector.
If you are in the jumper adding mood, I would add one to the reset line too. This could be useful if you have serial comms causing unwanted resets to the prop, if you do this it a pull up resistor must be put on the prop side, 10K should do. This is because the internal pull up is disabled when BOE is disabled..
The vast majority of my components are board pulls, I try to harvest as much as possible, from anywhere, SHE doesn't apriciate the "junk" but I am saving a fortune.
If only I could find an old board with a bunch of Props on it
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Why did I think a new, more challenging, job was a good idea ??
Pouring copper (connected to GND) in the free areas
Adding a jumper and a pull up resistor on the Reset track
Adding LEDs on the power lines
I read a few threads on the serial connection with the PC. Most of the time, only Rx and Tx lines are used (except for programming), probably because there is no interrupt procedure in the Prop.
So I would like to have your opinion about the other signals of the RS232 connection. Should I leave them unconnected or, for example, drive them to some spare pins of the 40 pins connector ? One of the piling up modules will be an Atmel Atmega. As this one uses +5V, it would perhaps be best not to use the Prop's RS232 connector to communicate with the PC but to allow its own RS232 connection. In this case I can leave the board as it is now.
To start with, I was thinking of designing a sort of Demo board with the 40 DIL version, adding 4 x female connector strip with 10 pin each (8 I/O ports + 3.3V + GND). Also possibly adding some dip switches to disconnect TV / VGA and Keyboard / Mouse for compatibility with Hydra board.
I am sure I am not the only one with this idea but I couldnt find such a homemade board through the whole forum . Could anyone show me a useful link ?
Jean-Marie
Jean-Marie,
Welcome. Take a look at the PROPELLER BREADBOARD PROJECT at: www.topprop.net (scroll down a bit on the page to find the project). I think that this may help you get started. You can build the thing on a regular solderless or soldered breadboard and then incorporate that into your own design.
Also George Herzog has created an excellent table outlining the ports used by the more popular (at that time) Prop systems. Old Bit Collector printed it in his excellent Propeller Cookbook. You can find that at http://ucontroller.com/Propeller%20Protoboard%20Designs%20for%20the%20Beginner.pdf. This may help you with selecting ports which are already covered by programing examples or complete programs.
It took me 2 days to find out why I could not access the forum with my password.
Thank you for your comments. I greatly appreciated the Propeller Cookbook.
I made a few new changes to the board, specially adapting the dimensions of the 40 pins connector (according to the real connectors I received yesterday) and adding 2 LEDS showing the traffic on the Rx and Tx lines (see the adapted schematic below).
I think I am not too far from the etching of the first board. Before that I will have to double check the layout.
I know I make things the wrong way but I am designing the schematic of the board. I will post it probably later today. It will help me to check the board.
Sorry for the spoof message, I thought I saw a problem on the reset jumper. Then I couldn't get onto the Forum and I was worried you would put copper to acid over that period. After posting these thoughts I looked again an saw that I was wrong and deleted the post.
Hello Toby, that's allright. The acid is not yet for today. There was a James Bond on TV.;)
Yes I think the jumper is at the right place. It can stop a reset from the serial but let you reset the Prop manually.
Below are some pdf files (They are of much better quality than the jpeg):
First the schematic. It combines the demo board and the serial programming.
Then the top side. One can see the components and the straps.
The third one is the bottom side : all the copper tracks.
The last one is everything together.
The problem I had with PCB software's inbuilt libraries is that they always assume that plated though holes are to be used and leave an absolute minimum of land area for the pads. This is great for giving plenty of room for track to pass through but not so great for the pads survival upon drilling.
I usually create my own pads so that, for 100mil pitch, the pad is 60mil wide leaving40 mil for a 20mil track and two 10mil gaps. For the 40 way header I would have 60 mil octagons if pass through tracks are used and 80mil ones if not. The 1mm drill for those pins is a bit close to 60 mil to start with and so I usually put a 0.6mm through as a pilot.
This is all with hand drilling with HSS bits, I suppose that a carbide bit in a machine station would be so much better.
My basic software doesn't give me much choice for the size of pads and tracks. The joined image shows the sizes I usually use when a track goes between pads. (I converted the metric sizes in mils)
For the holes I use 0.8 mm drill. If the hole has to be larger, I can take the oval pad (see picture) which is the same width as the round one.
Comments
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
If this is the case, I will have to redraw it with Eagle because my PCB design software is very basic and doesn't allow someone to round off the corners.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
The dictionary i use (www.wordreference.com/) only mention the hat of a bishop for miter or mitre (the French word is also "mitre").
Anyway, what do you think of this version ?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
At first I couldn't discern the red and green trace colors; I've done lots of boards (maybe I'm too color-blind) and use high contrast colors like red and blue, I couldn't tell you had a two sided board at first, so found lots of shorts. My bad eyes. Thread title threw me off; "Homemade singe side..."
There is still a 90° bent at upper-right. And an unconnected trace from the top pin of the regulator to the power jack.
Do you have any program to verify the layout connections vs. the schematic? Invaluable to find unconnected pins, shorted traces, traces too close, etc.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Harley Shanko
Thank you for examining the design so carefully. I have made the rectification.
No, my program doesn't make the verification with the schematic. Eagle does, but I don't like using Eagle because I find it so difficult to discover the right component within a huge library and I have never learnt to build my own library. Probably a bit too lazy !
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
Robert
This will be for tomorrow because this side of the earth is already in the darkness for a long time.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
There are two more 90°s; below pins 9 and 12 of the Prop. The ground one could be helped with a via and the trace from the left end of the big cap on the other side. I've done enough 4-layer board, power/ground on the inside and signal traces on outer sides for access visually and for mods if needed, so many of this sort of effect was absent.
I did one large 4-layer pcb, about 15 x 18", that would have been horrible without any open/short and proper continuity checking available, as it went through numerous iterations while being laid out (the designer wasn't finished yet the company head wanted the layout started way before the majority of the design placement was thought out by him. some 'fun', NOT!).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Harley Shanko
I added four .1µF caps near the EEPROM and the propeller. I hope Ididn't forget any other place.
______________________________________________________________________________
Hello Harley
I don't quite understand what your are saying. The board is single side. The red lines are the copper tracks underneath. The green lines are wires on the top. The kaki lines are the components or the edge of the board. I am sorry but I cannot change the colours.
So, I don't see what I can do to the 9O° angles below pin 9 and 12 of the Prop.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
The two 0.1uF caps you have added look to be very large. Either theu are can electrolytics or far too high a voltage rating. I use ceramic disk or surface mount ones (if you don't mind it being non-through hole). The disk ones are usually small enough to hide on the gaps of the IC base and the surface mount ones just fit across the tracks and do not present any thickness to the track side.
Many people, including me, try to get a additional 47-100uF Tantalum cap as close to the supply pins as possible, again there is usually space in the IC base.
I have fallen in love with the copper area pour, for a ground plain mostly. It has to help with shielding, and uses less Ferric Chloride too.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Why did I think a new, more challenging, job was a good idea ??
Post Edited (Toby Seckshund) : 8/3/2010 9:17:27 AM GMT
Time to etch the board?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.smarthome.viviti.com/propeller
Yes, the two .1µF caps between the Prop and the connector are electrolytic.
I have a box with all sorts of ceramic caps but I often have a problem to identify the values.
Could anyone help me to identify these three exemples ?
The cap 1 (204Z 12V) could be a .2µF 12V but I have no idea for the others
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
100 = 100pF
18 = 18pF
I'd use the 204.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
It is quite amazing. The 0.2µF is the smallest of the three caps but has the largest value ! It is 1 cm diameter and can go underneath the Prop.
By the way, I can't keep myself from asking another identification.
_______________________________________________________
@ Toby
I have to buy tantalum caps because I don't have any at home.
Unfortunately, my (basic) program doesn't pour copper areas, unless I make them myself by adding large lines one by one.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
Ooops! I had missed that point of the green color for the wires on pcb top.
Just ignore my comment about 90° trace bends. (As a friend used to say many times "First mistake".)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Harley Shanko
Lawson
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Lunch cures all problems! have you had lunch?
Post Edited (Lawson) : 8/3/2010 11:15:24 PM GMT
It is a good idea to have a jumper on the BOEN pin. I should add one.
Also I will have to check what is needed to use the RS232 for communication with an other controller, terminal or PC (other than programming purpose). Perhaps it would be worth forseeing one or two more jumpers so that I can use the same connector.
I leave now. I will not be back before tomorrow.
Thanks to all of you for your ideas.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
The vast majority of my components are board pulls, I try to harvest as much as possible, from anywhere, SHE doesn't apriciate the "junk" but I am saving a fortune.
If only I could find an old board with a bunch of Props on it
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Why did I think a new, more challenging, job was a good idea ??
I made a few changes to the layout:
I read a few threads on the serial connection with the PC. Most of the time, only Rx and Tx lines are used (except for programming), probably because there is no interrupt procedure in the Prop.
So I would like to have your opinion about the other signals of the RS232 connection. Should I leave them unconnected or, for example, drive them to some spare pins of the 40 pins connector ? One of the piling up modules will be an Atmel Atmega. As this one uses +5V, it would perhaps be best not to use the Prop's RS232 connector to communicate with the PC but to allow its own RS232 connection. In this case I can leave the board as it is now.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jean-Marie
Jean-Marie,
Welcome. Take a look at the PROPELLER BREADBOARD PROJECT at: www.topprop.net (scroll down a bit on the page to find the project). I think that this may help you get started. You can build the thing on a regular solderless or soldered breadboard and then incorporate that into your own design.
Also George Herzog has created an excellent table outlining the ports used by the more popular (at that time) Prop systems. Old Bit Collector printed it in his excellent Propeller Cookbook. You can find that at http://ucontroller.com/Propeller%20Protoboard%20Designs%20for%20the%20Beginner.pdf. This may help you with selecting ports which are already covered by programing examples or complete programs.
Good luck with your board.
It took me 2 days to find out why I could not access the forum with my password.
Thank you for your comments. I greatly appreciated the Propeller Cookbook.
I made a few new changes to the board, specially adapting the dimensions of the 40 pins connector (according to the real connectors I received yesterday) and adding 2 LEDS showing the traffic on the Rx and Tx lines (see the adapted schematic below).
I think I am not too far from the etching of the first board. Before that I will have to double check the layout.
Me too!
Your board looks very good. Time to get out the Ferric Chloride?
I know I make things the wrong way but I am designing the schematic of the board. I will post it probably later today. It will help me to check the board.
____________________________________________________
Hello Toby Seckshund
It is quite bizarre, I received a mail telling me that you had posted a reply but when I go on the forum, I don't see anything...
Sorry for the spoof message, I thought I saw a problem on the reset jumper. Then I couldn't get onto the Forum and I was worried you would put copper to acid over that period. After posting these thoughts I looked again an saw that I was wrong and deleted the post.
Yes I think the jumper is at the right place. It can stop a reset from the serial but let you reset the Prop manually.
Below are some pdf files (They are of much better quality than the jpeg):
First the schematic. It combines the demo board and the serial programming.
Then the top side. One can see the components and the straps.
The third one is the bottom side : all the copper tracks.
The last one is everything together.
I usually create my own pads so that, for 100mil pitch, the pad is 60mil wide leaving40 mil for a 20mil track and two 10mil gaps. For the 40 way header I would have 60 mil octagons if pass through tracks are used and 80mil ones if not. The 1mm drill for those pins is a bit close to 60 mil to start with and so I usually put a 0.6mm through as a pilot.
This is all with hand drilling with HSS bits, I suppose that a carbide bit in a machine station would be so much better.
My basic software doesn't give me much choice for the size of pads and tracks. The joined image shows the sizes I usually use when a track goes between pads. (I converted the metric sizes in mils)
For the holes I use 0.8 mm drill. If the hole has to be larger, I can take the oval pad (see picture) which is the same width as the round one.