| Author |
Message |
|
|
 Posted: Mar 03, 2006 - 10:45 PM |
|
Joined: May 26, 2004
Posts: 2538
Location: Las Vegas, Nevada
|
|
|
Quote:
there are photoelectricpickups on top of traffic signals that respond to coded flash sequences from strobe lights mounted on ambulances and fire trucks.
They're easy to make, but you can go to JAIL for 3 years for using one  Congress made it a Federal offence last year. If you feel reckless, you need about 20 infrared LEDs flashing at 10Hz. |
|
|
| |
|
|
|
|
|
|
Posted: Mar 04, 2006 - 02:35 AM |
|
Joined: Aug 04, 2005
Posts: 1267
|
|
I check and adjust the speed of the flash routines against the speed of the PC using the PC clock. It all seems to work then as long as the data rate is kept rather low...that is why this is good for changing settings for an AVR program but poor for doing something like implementing a bootloader.
I have thought about generating an avi file and then playing the flashing avi file back to send the data. Avi files seem to play back with really good timing accuracy regardless of what else the PC is doing.
If you implemented the avi idea and maximized the data rate and also had say 4 sections that were flashing data and 4 LEDs receiving that data on your device then speed could be really good  |
|
|
| |
|
|
|
|
|
|
Posted: Mar 04, 2006 - 02:53 AM |
|
Joined: Aug 04, 2005
Posts: 1267
|
|
Hi abcminiuser
That circuit cellar article was very useful..I had not even thought about their being a difference in the flashes from an LCD and a regular monitor
It seems I have much more work to do..I have been testing with only LCDs... |
|
|
| |
|
|
|
|
|
|
Posted: Mar 04, 2006 - 03:54 AM |
|
Joined: Nov 17, 2004
Posts: 6137
Location: Great Smokey Mountains.
|
|
The avi file is actually very good, I think they use DirectX or something to keep an area of the screen captive and it runs sort of independent of Windows. I'll check it out. Uh... eventually.
Smiley |
_________________
FREE TUTORIAL: 'Quick Start Guide for Using the WinAVR C Compiler with ATMEL's AVR Butterfly' AVAILABLE AT: http://www.smileymicros.com
|
| |
|
|
|
|
|
|
Posted: Mar 06, 2006 - 12:45 PM |
|
Joined: Dec 08, 2004
Posts: 4719
Location: Nova Scotia, Canada
|
|
|
Gwen wrote:
If you implemented the avi idea and maximized the data rate and also had say 4 sections that were flashing data and 4 LEDs receiving that data on your device then speed could be really good
If you were going to go down the multi-channel route, then why not go one step further and implement a synchronous serial algorithm...
Flash one section of the screen for the "clock", and then flash a second section (or more) with the "data" a la SPI.
That way you could output the data with a good probability of success no matter what else the PC might be doing. |
|
|
| |
|
|
|
|
|
|
Posted: Mar 06, 2006 - 01:32 PM |
|
Joined: Jun 27, 2005
Posts: 3412
Location: St.Petersburg, Russia
|
|
|
smileymicros wrote:
I've got a Timex watch with MicroSoft software on it that programs the watch by flashing the PC screen. I've often wondered how they control the timing on the flashes since Windows is indeterminate and can go off and do God-knows-what right in the middle of any time critical function you write. It is the exact opposite of a real-time system. So how do you plan to time the output pulses? I know it can be done, because my watch software does it, I just don't know how they do it.
NT (and subsequently, XP) have PerformanceCounters that have relatively high resolution (approx 10ms, which sucks ass but still better than nothing).
Check this article:
http://support.microsoft.com/kb/q172338/
And this Delphi-specific article:
http://www.delphifaq.com/faq/delphi_win ... f345.shtml
It's worth noting that despite their 10ms resolution, they don't guarrantee realtime performance, don't generate signals.. All in all useless except for maybe some kind of loop calibration. |
_________________
The Dark Boxes are coming.
|
| |
|
|
|
|
|
|
Posted: Mar 17, 2006 - 09:05 PM |
|
Joined: Mar 06, 2001
Posts: 552
Location: Moscow
|
|
Hi Zoom!
These is my waveform generator - GenerAll.
0-25MHz, +-3.0V.
Alexander |
|
|
| |
|
|
|
|
|
|
Posted: Mar 18, 2006 - 01:21 AM |
|
Joined: Nov 10, 2005
Posts: 1527
Location: Redmond, WA
|
|
Hi Alexander! Very nice! Are you using a DDS from Analog Devices?
-Zoom |
|
|
| |
|
|
|
|
|
|
Posted: Mar 18, 2006 - 09:10 AM |
|
Joined: Mar 06, 2001
Posts: 552
Location: Moscow
|
|
Hi Zoom!
Yes, it is the same you are using - AD9834.
ADM660 for -5V from +5V
AD8370 - for amplitude controll
AD8130 - for level shifting
AD8091 as buffer amp.
CP2102 for USB
Mega32
LM2621 as SEPIC
Alexander/ |
|
|
| |
|
|
|
|
|
|
Posted: Mar 18, 2006 - 10:52 PM |
|
Joined: May 25, 2004
Posts: 47
|
|
[quote]
That's a very nice LCD - very readable. Might I inquire where you got it?
Project image added to the gallery.
- Dean 
[/quote]
check out www.crystalfontz.com |
|
|
| |
|
|
|
|
|
|
Posted: Mar 19, 2006 - 12:15 PM |
|
Joined: Mar 06, 2001
Posts: 552
Location: Moscow
|
|
G'day mate!
I use MTC-S16208XFYHSGY-08 from Microtips Technology local distributor in Moscow http://www.cec-mc.ru.
Backlight is on on the foto.
At the work we are using OLED indicator now.
Alexander. |
|
|
| |
|
|
|
|
|
|
Posted: Mar 19, 2006 - 12:34 PM |
|
Joined: Oct 04, 2004
Posts: 442
Location: Turku, Finland
|
|
looks little bit diffrent then Alexander...
2- CH RF-transmitter |
|
|
| |
|
|
|
|
|
|
Posted: Mar 19, 2006 - 05:56 PM |
|
Joined: Aug 19, 2005
Posts: 1193
|
|
| Did you make the plastic shell yourself (or have it custom made)? Or did you find an old/existing remote and make a new circuit board to go inside? |
_________________
There's no place like ~/
|
| |
|
|
|
|
|
|
Posted: Mar 19, 2006 - 06:24 PM |
|
Joined: Oct 04, 2004
Posts: 442
Location: Turku, Finland
|
|
|
|
|
|
|
|
Posted: Mar 20, 2006 - 08:54 PM |
|
Joined: Feb 25, 2002
Posts: 807
|
|
Here's the original v1.0 prototype of the SuprAlarm. A configurable vehicle engine alarm that allows you to couple thresholds for alarm purposes.
For example, RPM=7000 is normally OK, but not if boost pressure is over 28lbs. If outside temp > 80F, max boost is 28lbs. If outside temp > 90F, max boost is 25lbs.
You can link up to 32 alarm conditions together. |
_________________
Dean 94TT
"Life is just one damn thing after another" Elbert Hubbard (1856 - 1915)
|
| |
|
|
|
|
|
|
Posted: Mar 20, 2006 - 09:00 PM |
|
Joined: Feb 25, 2002
Posts: 807
|
|
| And another one - this is actually in the Academy. A little guitar (bass and acoustic) tuner based on Jesper's one. |
_________________
Dean 94TT
"Life is just one damn thing after another" Elbert Hubbard (1856 - 1915)
|
| |
|
|
|
|
|
|
Posted: Mar 21, 2006 - 09:21 AM |
|
Joined: Jan 03, 2006
Posts: 4417
Location: Hemel Hemsptead, UK
|
|
|
|
|
|
|
|
Posted: Mar 21, 2006 - 04:25 PM |
|
Joined: Feb 25, 2002
Posts: 807
|
|
No - it doesn't talk OBD2 at all. It takes signals directly from various sensors already in the car, buffers and does gain/attenuation, then into the ADC inputs. It's a simple 2-line display, not a cool graphics one like yours. Then again, it's only a Mega8, not a 32, and it's full. There are a couple more things I would like to add, so I guess I'll switch it up to a Mega168.
It's main purpose is for racing, when the drive really doesn't have time to watch gauges all the time. You set thresholds, link them together, and if something goes out of whack, a LOUD alarm goes off. It does track what triggered it for later analysis.
The inputs are very flexible. Just set the gain resistors for the particular input, and there's a trimpot as well. In the description, R1 refers to R3, while R8 and R14 refer to R10 and R16. So for less than 5V, it's a direct shot into the opamp buffer. For more than 5V, it goes through a divider and the opamp buffers it. |
_________________
Dean 94TT
"Life is just one damn thing after another" Elbert Hubbard (1856 - 1915)
|
| |
|
|
|
|
|
|
Posted: Mar 21, 2006 - 07:23 PM |
|
Joined: Jan 03, 2006
Posts: 4417
Location: Hemel Hemsptead, UK
|
|
Yup, I can see why you'd want a range of input voltages... I'd be concerned about input impedances on things like lambda sensors (we've had occasional problems with air-fuel meters dragging the output voltage a little) but I was after a different approach - we have discovered that a lot of problems are related to dodgy sensors so it seemed logical to ask the ECU what it thought was happening. It's turned out to be a useful diagnostic tool.
Neil |
_________________
Neil Barnes
www.nailed-barnacle.co.uk
|
| |
|
|
|
|
|
|
Posted: Mar 21, 2006 - 07:40 PM |
|
Joined: Feb 25, 2002
Posts: 807
|
|
Yeah, I've always been concerned about the observer altering the observed. The 39k input resistor and opamp super-duper-high-impedance-inputs buffer help a lot. I've haven't seen any real corruption as yet, nor have any of my buddies running the SuprAlarm.
Heh, it saved a Supra last year Friend had one of the two fuel pumps fail in his 900RWHP Supra ... Normally the drop in fuel pressure wouldn't be too bad, but combined with 30lbs of boost would have been a catastrophic lean condition. When he got on it and boost climbed, the alarm went off, and the SuprAlarm pinpointed the cause. Me ? I'm only 440RWHP, I max out at 25lbs on the stock twins. Plenty quick though  |
_________________
Dean 94TT
"Life is just one damn thing after another" Elbert Hubbard (1856 - 1915)
|
| |
|
|
|
|
|
FAQ
Search
Register
Log in to check your private messages
Log in
Maximize
RSS
