Skip to main content ↑ ↓

Messages

This section allows you to view all Messages made by this member. Note that you can only see Messages made in areas you currently have access to.

Messages - arhi

46
General discussion / high speed low latency wireless connection between two fpga
hi guy's, brainstorming some project and I need to link two fpga chips without wires... I was considering wifi modules but all wifi modules I tried are fairly low (2mbps) on bandwidth and super slow (low latency, more then 10ms latency) .. anyone can suggest a good module for this? using fpga so fast spi or some parallel interface will do .. no need for slow uart these el-cheapo ones offer
52
Bus Pirate Support / Re: Buspirate v3 VS v4
yup, usb on mcp sucks big time :( .. didn't know you managed to solve the pps restart issues .. cool
arm is too expensive unless you go cm0 with some prc manufacturer but then you have to purchase gzilion chips and support is zero so really not sure if that's the way to go ..

as for the usb part, mcp is shooting their usb2serial that's 18f something .. you can get usb capable pic for peanuts, way cheaper then ftdi and since you control firmware (even if you go with semiclosed mcp usb lib) you can get waaaaaaaaay more out of it then from ftdi ... I'd go with parallel communication, 8bit or 4bit depends on the chips .. for e.g. PIC16F1454 is ~1$, has USB and enough pins .. even if you go with SPI it can do 12MHz SPI!!! that's 1.4 megabytes per second (11mbps so basically just as fast as usb)!!! I doubt you need faster then that :D .. and for 1$ :) it's imo way better then any ftdi .. on the business side of the device .. well, that you have more experience for bus pirate then I do .. depends on what you want to achieve .. I'd personally go with XMOS just for kicks of it, it's great and interesting chip :D .. I would not go with arm .. I'd go either with XMOS + external flash or some pic with enough flash to put a lot of code in I can call from scripts.. it's really all about how you want to improve on 3b and what additional features you want .. XMOS is faster and more versatile then most of chips you can put there... it's on the other hand not nearly as easy to use as pic
53
Bus Pirate Support / Re: Buspirate v3 VS v4
so v4 is dead ... as expected, bp uses pps to get needed hw on same output pins and to change pin setup you have to reset, when you reset you kill usb connection ... has nothing to do with usb on pic but with design decision and way pps works ..

v5 will be ftdi again? maybe use some cheaper pic instead of ftdi for usb, then parallel connection to another pic for the work .. you get both fast transfers, full control and ability to use pps they way it works on v3 just with more flash and more speed and greater umpf
56
Project logs / Re: HAKKO (907ESD) and SOLOMON (SL-10/30) soldering iron dri
I did not sell mine, gave excess ones already so have no more to give. The DP version is available on seeedstudio (same schematic as mine pcb only all components are trough hole so lot easier to solder, mine has lot of smd components cramped together with very ugly layout and even uglier silk screen)
57
Project logs / Re: HAKKO (907ESD) and SOLOMON (SL-10/30) soldering iron dri
[quote author="TomKeddie"]
- R30 is 0.0001R ?
- R31 is 1000M ?
[/quote]

This depends on
1. do you use NTC1/NTC2 at all? Firmware does not use them at all. I just short R30 (proteus didn't allow me to enter 0 for the value) and leave R31 open and U2:A behaves as buffer. If you need amplification you use those resistors

[quote author="TomKeddie"]
- any specifics on the triac or optocoupler?
[/quote]
 any triac that can handle 30vac and 100w should do, personally i never seen one that is below that

R37,R38, U6, D9, Q2 are only needed for DC driver if you power it from ac these parts you do not need

[quote author="TomKeddie"]
- what voltage is the lm2574?
[/quote]

5v

[quote author="TomKeddie"]
- what voltage is the LM385?  5.0V seems to be missing from mouser.  I guess 78L05 will work?
[/quote]
no, 385 is reference, 78l05 is crappy regulator
use lm385 2.5v or 1.25v. if you use 2.5v cal1 need to ne 5k, if you use 1.25v cal1 need to be 2k

[quote author="TomKeddie"]
- what package are the inductors?
[/quote]
what can fit on the pcb :)
make sure you have indictors rated for min 1A


[quote author="TomKeddie"]
- what frequency is the crystal?
[/quote]
20mhz
58
Project development, ideas, and suggestions / Re: Dangerous Soldering Iron driver V1.5(based on Arhi's gen
I don't think it's the negative voltage on contrast problem as if you look at video
- in the beginning he sees squares in one line (so his contrast is more/less ok)
~7sec he restart it - blank screen (ok)
~14sec he press some button and screen start to dance ?!?!?!

that is garbage going into lcd, the initialization did not go properly and then data that is sent to lcd is making the weird waving .. first thing to do is set those lcd_delay things to slower, if that does not work few other things could be tried (the strobe_e macro)
59
Project development, ideas, and suggestions / Re: Dangerous Soldering Iron driver V1.5(based on Arhi's gen
lot of buttons :D
that video is weird :(
the "boxes" you get before lcd is initialised, when they are gone it looks like lcd got initi string but then the "dancing" that's strange. I see only few wires going to lcd, did you ground the unused data lines on lcd (d0-d3)? How fast is that lcd? did you maybe try different brand? The firmware is running with RW pin grounded so I'm not testing busy flag but I have a delay. I did that because I designed this to go with some COG display's I have that does not support busy flag (their RW pin is N/C) but maybe the delay in firmware is too short for your lcd. Increasing delay would be easiest way to go. If you can compile the firmware yourself, get it from http://sourceforge.net/p/arhimed-sid/code/HEAD/tree/ and edit LCD.c changing
Code: [Select]
/* SAFE
#define LCD_DELAY_LONG() Delay10KTCYx(48) //delay_ms(40)
#define LCD_DELAY() Delay10KTCYx(19) //delay_ms(16)
#define LCD_DELAY_F() Delay10KTCYx(5) //delay_ms(4)
#define LCD_DELAY_SHORT() Delay10TCYx(5) //delay_us(4)
*/
#define LCD_DELAY_LONG() Delay10KTCYx(48)
#define LCD_DELAY() Delay10KTCYx(1)
#define LCD_DELAY_F() Delay10KTCYx(20)
#define LCD_DELAY_SHORT() Delay10TCYx(5)

to:

Code: [Select]

#define LCD_DELAY_LONG() Delay10KTCYx(48) //delay_ms(40)
#define LCD_DELAY() Delay10KTCYx(19) //delay_ms(16)
#define LCD_DELAY_F() Delay10KTCYx(5) //delay_ms(4)
#define LCD_DELAY_SHORT() Delay10TCYx(5) //delay_us(4)
/* UNSAFE but fast
#define LCD_DELAY_LONG() Delay10KTCYx(48)
#define LCD_DELAY() Delay10KTCYx(1)
#define LCD_DELAY_F() Delay10KTCYx(20)
#define LCD_DELAY_SHORT() Delay10TCYx(5) 
*/

recompile and you should have it working with slow lcd's.
You can even make LCD_DELAY() longer if this is still too fast for your lcd
60
Project development, ideas, and suggestions / Re: Dangerous Soldering Iron driver V1.5(based on Arhi's gen
[quote author="Aleksej"]Hi everybody,

I just finished soldering everything together and I have a problem.
When I turn it on, pwr led starts and display backlight starts but I don't see almost anything else on the display (I get couple of some strange dots in the beginning but they disappear after couple of seconds). Where should I start looking?
I didn't connected any soldering iron to the board yet.[/quote]

follow the pre-calibration procedure before you get any iron attached.
check the lcd contrast pot you should be able to see tekst.
if there is nothing displayed on the lcd even if you turn pot max to min then

0. check 5V rail going to 18f2550, see what voltage you have there and if you have a scope check ripple there too
1. check if you can read back what you wrote to the 18f2550 to see if you programmed it correctly
2. check traces to xtal and loading caps
3. check decoupling cap on the 18f2550
4. check traces and solder points from 18f2550 to lcd

( ! ) Fatal error: Uncaught exception 'Elk_Exception' with message 'Please try again. If you come back to this error screen, report the error to an administrator.' in /var/www/dangerousprototypes/forum/sources/database/Db-mysql.class.php on line 696
( ! ) Elk_Exception: Please try again. If you come back to this error screen, report the error to an administrator. in /var/www/dangerousprototypes/forum/sources/database/Db-mysql.class.php on line 696
Call Stack
#TimeMemoryFunctionLocation
10.01742452888session_write_close ( )...(null):0
20.01772584504ElkArte\sources\subs\SessionHandler\DatabaseHandler->write( )...(null):0
30.01772585280Database_MySQL->query( ).../DatabaseHandler.php:119
40.06032724040Database_MySQL->error( ).../Db-mysql.class.php:273