Rockbox mail archive
Subject: Re: BMW IBUS <-> Archos Jukebox (as CD Changer)
From: Fernando Birra (fpb_at_di.fct.unl.pt)
"[IDC]Dragon" <idc-dragon_at_gmx.de> escreveu na mensagem
>> In my case if I connect the archos directly to the bus I read around 8V
>> the pin/bus when the archos forces the bus to low state.
>Uh, oh. It should collapse to about 4V when you connect to an open drain /
>open collector bus. If you get more, either some bus pullup is way to
>(impedance too low), or something is _driving_ the bus high (then it
>be open drain). Very harmful to your poor remote input!
>If you really have 8V on the remote pin, the RX pin of your CPU is in
Well, perhaps it isn't really 8V. I'm not sure at this moment and really,
I'm afraid to try again...
>> I tried changing and it still did't work. Also, doesn't the fact that the
>> direct connection (without any level shifter) also has the problem of not
>> being capable of driving the bus low suggests that the problem are not
>Yes, definitely. Is your connector OK? Gnd is the outermost pin, RX is the
>second. Camcorder TV out cables I've seen have these pins in reverse order.
Yes, the Camcorder cable screws everything but I did verify that in advance.
Presently I'm just using one RCA connector (Ground+PB10 pin).
As I said, transmission and reception works well in my simulated bus and
reception also works in the car...
>>>Another explaination could be that the pullup is very strong, so the
>>>fails to pull it low. How much current do you measure when you short
>>>to ground (or bus to ground)? The SH is specified to sink max. 10 mA.
I think this is the problem. When I tried to measure the necessary current
to bring the bus to 0V (by shorting IBUS and GND with the multimeter) my
multimeter displayed 0.2 in the 200mA range. However when I switched to the
20mA range it displayed 0.02. So I'm not trusting the readings. However, the
PB10 pin of the archos is at 1.2V and that (by the datasheet) suggests that
the archos is sinking as much current as it can (10mA). So, I think this is
really the problem.
If my assumption is correct, what can be done to my circuit? I'm using the
one pictured here: http://ctp.di.fct.unl.pt/~fpb/ibus-archos.jpg as it seems
to be the one that works the best...
Since the 1.2V is not enough to make the IBUS go low (I guess it is a
problem of the Vgs threshold) should I replace the MOSFET with another one
with a lower threshold? Or is there a way to limit the current that flows
into the archos pin so that it can bring the pin to a lower voltage?
0.4V-0.6V would be nice I guess... Should I decrease the value of the
resistor placed between the two MOSFETs? Wouldn't that increase the current
that flows through there when the bus is in low state?
>> I'm not sure if I understood it (I really know very little about
>> electronics). Do you want me to measure the current that through to the
>> source or drain pins when either the source or bus is short to ground? I
>> tried but I always got 0 mA, I'm sure I made some mistake...
>> Can I connect one test lead of the multimeter to the ground and the other
>> one to the source pin to measure it?
>Use your multimeter in maybe 200 mA range setting to short the bus to
>ground. For amp measurements, most meters require to move the positive
>measure in parallel to a short you already made (you don't need one),
>then the current travels there and not through your meter.
OK, I know that part. Sure, the current needs to flow through the amp meter
and not through the circuit ;)
I really don't trust the amp readings... They are not consistent between
>Let's hope for the best.
Thanks for all
Page was last modified "Jan 10 2012" The Rockbox Crew