usb problems

[Gene Heskett]

On Sunday 12 October 2014 08:38:36 Colin Law did opine
And Gene did reply:
> On 12 October 2014 13:03, Gene Heskett <gheskett at wdtv.com> wrote:
> > On Sunday 12 October 2014 07:37:21 Gene Heskett did opine
> > 
> > And Gene did reply:
> >> On Sunday 12 October 2014 02:41:15 Nils Kassube did opine
> >> 
> >> And Gene did reply:
> >> > Gene Heskett wrote:
> >> > > I am in need of a tool that will effectively do a ping against a
> >> > > specified USB port, a port I know there is something on the
> >> > > other end of the cable and hub as I just had a printer do a
> >> > > test page over the same cabling. Basically I want to send a $55
> >> > > down the cable once a second to a specified port, in thgius
> >> > > case /dev/ttyUSB1, for oscilloscope troubleshooting, at least
> >> > > as long as trouble doesn't shoot back... ;-)
> >> > 
> >> > If that /dev/ttyUSB1 device is a USB->serial converter, you could
> >> > treat it like any terminal device. Therefore, you can use stty to
> >> > disable the handshaking lines and set the desired speed (e.g. 100
> >> > Baud). Then echo the desired characters to the device:
> >> > 
> >> > stty -F /dev/ttyUSB1 clocal ospeed 100
> >> 
> >> stty -F /dev/ttyUSB1 clocal ospeed 115200
> >> stty -F /dev/ttyUSB1 clocal ispeed 115200
> >> while true;do echo -n "e" >/dev/ttyUSB1;sleep 1;done
> >> 
> >> > You can cancel the loop with CTRL-C.
> >> > 
> >> > 
> >> > Nils
> >> 
> >> This works, I can see the once a second red led flicker.  But when
> >> measured with the scope, the baud rate seems to be about 55800 (not
> >> 57600) unless that "e" is a 2 bits high, 2 bits low etc character.
> >> 
> >> Cheers, Gene Heskett
> > 
> > and after trying to use it at 115200, something is resetting it to
> > 9600?
> > 
> > The .jar that makes a server out of this machine has a config, and
> > that is all correct.  Showing:
> > Sending command: dw config set DeviceType serial        OK
> > Sending command: dw config set SerialDevice /dev/ttyUSB1        OK
> > Sending command: dw config set SerialRate 115200        OK
> > Sending command: dw config set SerialParity none        OK
> > Sending command: dw config set SerialStopbits 1 OK
> > Sending command: dw config set SerialDTR false  OK
> > Sending command: dw config set SerialRTS false  OK
> > Sending command: dw config set SerialFlowControl_RTSCTS_IN false     
> >   OK Sending command: dw config set SerialFlowControl_RTSCTS_OUT
> > false       OK Sending command: dw config set [@name] CoCo 3 on
> > /dev/ttyUSB1   OK Sending command: dw config set [@desc] Autocreated
> > 2014-10-12 07:30:40.164 OK
> > Sending command: dw config set UseMIDI true     OK
> > Sending command: dw config set CurrentPrinter Text      OK
> > Sending command: dw config set Printer(1).OutputDir /CoCo/dwcache    
> >   OK
> > 
> > However, any attempt from the CoCo3 to do a directory listing is a
> > read error because no data is coming back, the request for that data
> > never arrived at this machines USB1 port.
> > 
> > I've taken the serial cable apart, its well soldered on both ends,
> > and I've tried the 2nd of 2 identical converters, converters that
> > worked just fine in April of this year.  Data is coming out of the
> > CoCo3 with about a 10 volt p-p swing centered on ground.  Humm, come
> > to think about that, thats low, that circuit is running on the
> > output of an old XT psu, and all voltages there are good, +-12 is at
> > least +-11.5 volts.  I've blown another S.A.L.T chip would seem to
> > be the correct diagnosis.
> > 
> > This is a jury rigged external power supply as the original was +-8
> > volts. The chip has been replaced once and I have one more spare. 
> > Unfortunately, the data available on this chip is limited, and
> > before I blow another, I'll order me up a set of 7808 and 7908's to
> > make that +_12 look like what its used to.
> 
> RS232 specifies a min of +-3 volts up to +-15 if I remember correctly,
> so +-5 should be fine.

And is rated to take surges to at least 25 volts, usually more from most 
chip vendors.

But is it enough to power the internal circuitry?  Or do these converters 
have built in charge pumps running on the +4.5 or so they can get from the 
USB port?  Its plugged into a powered hub, and its blue power led is the 
usual brightness.

The converter itself is:

T:  Bus=02 Lev=05 Prnt=17 Port=03 Cnt=01 Dev#= 19 Spd=12  MxCh= 0
D:  Ver= 1.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs=  1
P:  Vendor=0403 ProdID=6001 Rev=04.00
S:  Manufacturer=FTDI
S:  Product=USB HS SERIAL CONVERTER
S:  SerialNumber=FTDHG6E2
C:  #Ifs= 1 Cfg#= 1 Atr=a0 MxPwr=44mA
I:  If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=ftdi_sio

Once the house is up for the day, I'll make coffee to prop my eyes open 
better, drag my soldering station in, or take the motherboard to the shop, 
and fix the trace cut that left that chip with only 5 volts+, superglue a 
small heat sink to the top of my last chip and try it again.  It did work 
w/o the heat sink for years, but I concocted a feedback loop in the 
dw4servers config last spring which was shipping an error signal back and 
forth at full 115kb bandwidth, which overheated the driver in this chip 
and crowbarred it eventually.

> Colin

Thanks Colin.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS