LinkBack URL
About LinkBacks
Reply With QuoteFor testing you can use normal 2.8mm fastons (non insulated, like theseOriginally Posted by Ale
, note that the picture is the 4.8mm one) with some tape to prevent shorts.
I didn't test with the insulated one, I don't think it will fit.
See my change in schematics at post #205.
For testing you can use normal 2.8mm fastons (non insulated, like these
I didn't test with the insulated one
Not today, maybe tomorrow.
And the answer is : pretty well :-)
So I just threw it all away and only left the two 10k resistors inline between the Rx/Tx of the HU and the serial ports, and all looks well. The only glitch I had was while powering up the Head Unit : the first byte I received was a 0x00000000, so I guess a spike came along the line and was considered a start bit, then 8 data bits were all at zero. The parser should take care of discarding that.
And indeed, the fact that the input impedance is between 3K and 7K is part of the RS232 specification (see http://www.rs485.com/rs485spec.html for example), so putting a 10K resistor in series actually creates a 3:1 divider which makes sure "low" voltage is under the input threshold... but a 5K or even lower could most probably work
I'll try to wire the active part tomorrow.
After my few tests with open wires, I see that the original CDC doesn't want to go further if its "BOOTING" messages are not acknowledged. Conversely, maybe the HU doesn't want to acknowledge anything until it has received this first "BOOTING" message...I just repeated without successmy test with no interface (just a divider network on input and a diode on output, I tried even without a diode).
So what you could try, instead of sending "PLAYING" frames, would be to send "BOOTING" frames and see if they get acknowledged. As they are the first frames sent, you can send the following fixed bytes and see if you get an answer :
3Dh 01h 03h 11h 60h 06h 48h
My tests show that they get repeated by the CDC (without the frame ID being incremented) until he gets an acknowledge, and that the HU ackowledges them all (once again, even though they have the same frame ID), so that looks like a bulletproof test to do.
That's what my test program (in post #132, though I made some changes since then) tries to do, alas without receiving an acknowledgment, so I cannot go further. Mmmh I actually send 00 as sequence, I'll try with 01.
After all i'am thinking that the best solution is to use a chip made for rs232, like MC145406 it has already pullup and pulldown resistor, ESD protection etc. We supply only VCC +5V for the digital section (78L05 may be) and VDD+12V from the HU, the negative VSS not used and tied to GND, so we can use the cascaded inverter configuration used till now in the spy circuit, for both direction PCTX->HURX and HUTX->PCRX. Max3232 can't be used becouse of its max V+ of only 7V.
A mere glimpse of all the schematics we've seen passing by.
The upcoming weekend appears to be a good time for a trip to Brussels, if that is convenient for you. Sunday, to be more specific, but subject to change. We'll exchange details by PM later this week.
CarPC status: HW all done, SW needs tweaked.
Hardware: VIA MII-12K, 512MB, 60GB 2.5", CW-8123 DVD-CDRW, 7" Lilli ts, Opus 90W, BU-353 GPS, 802.11b PCI, USB bluetooth dongle, AverMedia AverTV Cardbus Plus, Morex Cubid 3677
Software: RR, MM/FD
Agreed.
Simpler : we have the PC's 5V :-)like MC145406 it has already pullup and pulldown resistor, ESD protection etc. We supply only VCC +5V for the digital section (78L05 may be)
...or PC too :-)and VDD+12V from the HU
Yes.the negative VSS not used and tied to GND, so we can use the cascaded inverter configuration used till now in the spy circuit, for both direction PCTX->HURX and HUTX->PCRX. Max3232 can't be used becouse of its max V+ of only 7V.
Although, I asked my local dealer and he has a 15-days delay to order it (and it's 4.6 EUR a unit).
So I'd try to do something with MAX3232 and a few transistors like you proposed in post 205.
Indeed.A mere glimpse of all the schematics we've seen passing by.
By the way, I just thought about something (for the only one of us who has a CDC :-)). I think I'll leave the "reading" part always connected so that even when on CDC position, I can do a special combination (e.g. press pause twice quickly) that would make the PC take the control or release it...
Sunday would be fine indeed.The upcoming weekend appears to be a good time for a trip to Brussels, if that is convenient for you. Sunday, to be more specific, but subject to change. We'll exchange details by PM later this week.
Posting Permissions
Bookmarks