DCF77, USB <-> parallel adapter, any solution?

classic Classic list List threaded Threaded
25 messages Options
12
Reply | Threaded
Open this post in threaded view
|

DCF77, USB <-> parallel adapter, any solution?

Peter Maffter
I was using a DCF77 Dongle on the parallel port on my older machine and got the DCF77 time information in Europe.
This worked quite fine most of the time.

My new machine has no parallel port.
I tried several USB <-> Parallel adapters (like the DeLock USB910) but they all failed.
Some people posted on the net that these are only printer adapters.

Are there any USB <-> Parallel adapters that work with dongles, non printer devices like the
Conrad DCF77 ( see http://man.sourcentral.org/SUSE101/4+pcfclock ,

picture: http://www.linum.com/typo3temp/GB/499b753382.jpg?file=uploads%2Fpics%2Ffunkuhr-conrad-lpt-540-20-20.jpg&md5=fdd0be05be243392bdb490c7925160d3e0e46f85&parameters[0]=YTo0OntzOjU6IndpZHRoIjtzOjM6IjgwMCI7czo2OiJoZWlnaHQiO3M6NDoiNjAw&parameters[1]=bSI7czo3OiJib2R5VGFnIjtzOjU4OiI8Ym9keSBzdHlsZT0iYmFja2dyb3VuZDoj&parameters[2]=RkZGOyBtYXJnaW46IDBweDsgcGFkZGluZzogMHB4OyI%2BIjtzOjQ6IndyYXAiO3M6&parameters[3]=Mzc6IjxhIGhyZWY9ImphdmFzY3JpcHQ6Y2xvc2UoKTsiPiB8IDwvYT4iO30%3D )

Any other solutions that have DCF77 or GPS time receiving functionality even if the machine is turned off?


Thanks in advance

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
Peter Maffter wrote:

> I was using a DCF77 Dongle on the parallel port on my older machine
> and got the DCF77 time information in Europe. This worked quite fine
> most of the time.
>
> My new machine has no parallel port.
> I tried several USB <-> Parallel adapters (like the DeLock USB910) but
> they all failed. Some people posted on the net that these are only
> printer adapters.
>
> Are there any USB <-> Parallel adapters that work with dongles, non
> printer devices like the Conrad DCF77 ( see
> http://man.sourcentral.org/SUSE101/4+pcfclock ,
>
> picture:
>
http://www.linum.com/typo3temp/GB/499b753382.jpg?file=uploads%2Fpics%2Ffunkuhr-conrad-lpt-540-20-20.jpg&md5=fdd0be05be243392bdb490c7925160d3e0e46f85&parameters[0]=YTo0OntzOjU6IndpZHRoIjtzOjM6IjgwMCI7czo2OiJoZWlnaHQiO3M6NDoiNjAw&parameters[1]=bSI7czo3OiJib2R5VGFnIjtzOjU4OiI8Ym9keSBzdHlsZT0iYmFja2dyb3VuZDoj&parameters[2]=RkZGOyBtYXJnaW46IDBweDsgcGFkZGluZzogMHB4OyI%2BIjtzOjQ6IndyYXAiO3M6&parameters[3]=Mzc6IjxhIGhyZWY9ImphdmFzY3JpcHQ6Y2xvc2UoKTsiPiB8IDwvYT4iO30%3D
> )
>
> Any other solutions that have DCF77 or GPS time receiving
> functionality even if the machine is turned off?

Receiving is no problem, but if you want the time displayed like on your
parallel thingie, I don't know.  We use a Tobit serial DCF77 receiver.


--
Per Jessen, Zürich (15.2°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Roger Oberholtzer
On Tue, 2012-08-28 at 07:36 +0200, Per Jessen wrote:
> Tobit serial DCF77 receiver

OOC, how accurate are these? The signal has the resolution if one
second. But I guess when that second is, is very accurate, since it is
based on the German national atomic clock. There is a delay (3
milliseconds at 1000 km from Frankfurt). But other than that, any idea
how accurate your PC clock is when using this to set the time?


Yours sincerely,

Roger Oberholtzer

OPQ Systems / Ramböll RST

Office: Int +46 10-615 60 20
Mobile: Int +46 70-815 1696
[hidden email]
________________________________________

Ramböll Sverige AB
Krukmakargatan 21
P.O. Box 17009
SE-104 62 Stockholm, Sweden
www.rambollrst.se


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
Roger Oberholtzer wrote:

> On Tue, 2012-08-28 at 07:36 +0200, Per Jessen wrote:
>> Tobit serial DCF77 receiver
>
> OOC, how accurate are these?

As accurate as the DCF77 signal :-)

> The signal has the resolution if one second. But I guess when that
> second is, is very accurate, since it is based on the German national
> atomic clock.

Right.

> There is a delay (3 milliseconds at 1000 km from Frankfurt). But other
> than that, any idea how accurate your PC clock is when using this to
> set the time?

NTP knows about transmission delays and all that - to be honest, I don't
know how accurate it is, I have nothing to compare to.  I am not
concerned about milliseconds myself, but I suspect that is the level of
accuracy achievable.



--
Per Jessen, Zürich (19.7°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Roger Oberholtzer
On Tue, 2012-08-28 at 10:14 +0200, Per Jessen wrote:

> NTP knows about transmission delays and all that - to be honest, I don't
> know how accurate it is, I have nothing to compare to.  I am not
> concerned about milliseconds myself, but I suspect that is the level of
> accuracy achievable.

I have read that inexpensive receivers, like in consumer grade clocks,
should be able to achieve 0.1 sec accuracy. Better receivers can do
better.

But, returning to the thread, would this be the case when a USB adapter
is added? I know that such solutions do not work with serial port GPS
that provide a PPS, resulting in far less accurate time synchronization.
Maybe the parallel or serial port versions of these devices require
access to a physical interrupt line that is not possible when USB is
used?


Yours sincerely,

Roger Oberholtzer

OPQ Systems / Ramböll RST

Office: Int +46 10-615 60 20
Mobile: Int +46 70-815 1696
[hidden email]
________________________________________

Ramböll Sverige AB
Krukmakargatan 21
P.O. Box 17009
SE-104 62 Stockholm, Sweden
www.rambollrst.se


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: EXTERNAL: [opensuse] DCF77, USB <-> parallel adapter, any solution?

Damon Register
In reply to this post by Peter Maffter
On 8/27/2012 3:52 PM, Peter Maffter wrote:
> My new machine has no parallel port.
> I tried several USB <-> Parallel adapters (like the DeLock USB910) but they all failed.
> Some people posted on the net that these are only printer adapters.
Is an internal parallel card an option?

Damon Register

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

James Knott
In reply to this post by Per Jessen
Per Jessen wrote:
> NTP knows about transmission delays and all that - to be honest, I don't
> know how accurate it is, I have nothing to compare to.  I am not
> concerned about milliseconds myself, but I suspect that is the level of
> accuracy achievable.

Is NTP over the Internet not suitable?
--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
In reply to this post by Roger Oberholtzer
Roger Oberholtzer wrote:

> On Tue, 2012-08-28 at 10:14 +0200, Per Jessen wrote:
>
>> NTP knows about transmission delays and all that - to be honest, I
>> don't know how accurate it is, I have nothing to compare to.  I am
>> not concerned about milliseconds myself, but I suspect that is the
>> level of accuracy achievable.
>
> I have read that inexpensive receivers, like in consumer grade clocks,
> should be able to achieve 0.1 sec accuracy. Better receivers can do
> better.

I could be way off, but I don't think it is related to the receiver
quality. After all, either you've got a signal or you don't.  What is
done with the data received is probably more important, and that's
where NTP comes in.

> But, returning to the thread, would this be the case when a USB
> adapter is added? I know that such solutions do not work with serial
> port GPS that provide a PPS, resulting in far less accurate time
> synchronization. Maybe the parallel or serial port versions of these
> devices require access to a physical interrupt line that is not
> possible when USB is used?

Using a PPS is a different solution, but I have also seen people having
trouble with USB and PPS sources.



--
Per Jessen, Zürich (24.7°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
In reply to this post by James Knott
James Knott wrote:

> Per Jessen wrote:
>> NTP knows about transmission delays and all that - to be honest, I
>> don't know how accurate it is, I have nothing to compare to.  I am
>> not concerned about milliseconds myself, but I suspect that is the
>> level of accuracy achievable.
>
> Is NTP over the Internet not suitable?

Yes, absolutely but as I had the DCF77 receiver anyway I went with that.
I have in fact two internet stratum 2 peers configured too.  


--
Per Jessen, Zürich (24.8°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

James Knott
Per Jessen wrote:
> Yes, absolutely but as I had the DCF77 receiver anyway I went with that.
> I have in fact two internet stratum 2 peers configured too.

Well, if you still want to use that receiver, then I guess your only
option is to get a parallel port card for your computer.  I think I have
one here for the old ISA bus.  ;-)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
James Knott wrote:

> Per Jessen wrote:
>> Yes, absolutely but as I had the DCF77 receiver anyway I went with
>> that. I have in fact two internet stratum 2 peers configured too.
>
> Well, if you still want to use that receiver, then I guess your only
> option is to get a parallel port card for your computer.  I think I
> have one here for the old ISA bus.  ;-)
>

The OP had a problem with a USB parallel port, my old fashioned serial
setup is working fine :-)  


--
Per Jessen, Zürich (24.5°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Roger Oberholtzer
In reply to this post by Per Jessen
On Tue, 2012-08-28 at 14:12 +0200, Per Jessen wrote:

> Roger Oberholtzer wrote:
>
> > On Tue, 2012-08-28 at 10:14 +0200, Per Jessen wrote:
> >
> >> NTP knows about transmission delays and all that - to be honest, I
> >> don't know how accurate it is, I have nothing to compare to.  I am
> >> not concerned about milliseconds myself, but I suspect that is the
> >> level of accuracy achievable.
> >
> > I have read that inexpensive receivers, like in consumer grade clocks,
> > should be able to achieve 0.1 sec accuracy. Better receivers can do
> > better.
>
> I could be way off, but I don't think it is related to the receiver
> quality. After all, either you've got a signal or you don't.  What is
> done with the data received is probably more important, and that's
> where NTP comes in.

The DCF77 wikipedia entry (correctly?) states that inexpensive receivers
typically have a 10 Hz bandwidth, and thus have this limitation. Not
sure what that means. For this kind of sync, it is not the frequency of
the update, but the accuracy of it when it arrives. This is a bit beyond
my expertise.

> > But, returning to the thread, would this be the case when a USB
> > adapter is added? I know that such solutions do not work with serial
> > port GPS that provide a PPS, resulting in far less accurate time
> > synchronization. Maybe the parallel or serial port versions of these
> > devices require access to a physical interrupt line that is not
> > possible when USB is used?
>
> Using a PPS is a different solution, but I have also seen people having
> trouble with USB and PPS sources.

By 'Trouble' you mean it does not work. USB cannot propagate the PPS
signal with the same accuracy as the serial port. Some don't even look
at it. I checked the kernel drivers a while back to see what each USB
serial adapter supported. They are geared more to tty control for things
like modems.


Yours sincerely,

Roger Oberholtzer

OPQ Systems / Ramböll RST

Office: Int +46 10-615 60 20
Mobile: Int +46 70-815 1696
[hidden email]
________________________________________

Ramböll Sverige AB
Krukmakargatan 21
P.O. Box 17009
SE-104 62 Stockholm, Sweden
www.rambollrst.se


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
Roger Oberholtzer wrote:

> On Tue, 2012-08-28 at 14:12 +0200, Per Jessen wrote:
>> Roger Oberholtzer wrote:
>>
>> > On Tue, 2012-08-28 at 10:14 +0200, Per Jessen wrote:
>> >
>> >> NTP knows about transmission delays and all that - to be honest, I
>> >> don't know how accurate it is, I have nothing to compare to.  I am
>> >> not concerned about milliseconds myself, but I suspect that is the
>> >> level of accuracy achievable.
>> >
>> > I have read that inexpensive receivers, like in consumer grade
>> > clocks, should be able to achieve 0.1 sec accuracy. Better
>> > receivers can do better.
>>
>> I could be way off, but I don't think it is related to the receiver
>> quality. After all, either you've got a signal or you don't.  What is
>> done with the data received is probably more important, and that's
>> where NTP comes in.
>
> The DCF77 wikipedia entry (correctly?) states that inexpensive
> receivers typically have a 10 Hz bandwidth, and thus have this
> limitation.

Did you leave out a 'k'?  I'm sure the bandwidth is more like 10kHz, but
I think that's only of important if you're using the carrier to
generate a PPS. See below.

> Not sure what that means. For this kind of sync, it is not
> the frequency of the update, but the accuracy of it when it arrives.
> This is a bit beyond my expertise.

The 77.5kHz longwave carrier is also highly accurate and can be used to
generate a very good PPS signal.

>> Using a PPS is a different solution, but I have also seen people
>> having trouble with USB and PPS sources.
>
> By 'Trouble' you mean it does not work. USB cannot propagate the PPS
> signal with the same accuracy as the serial port.

Yup.


--
Per Jessen, Zürich (24.6°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Roger Oberholtzer
On Tue, 2012-08-28 at 16:05 +0200, Per Jessen wrote:
 
> > The DCF77 wikipedia entry (correctly?) states that inexpensive
> > receivers typically have a 10 Hz bandwidth, and thus have this
> > limitation.
>
> Did you leave out a 'k'?  I'm sure the bandwidth is more like 10kHz, but
> I think that's only of important if you're using the carrier to
> generate a PPS. See below.

10 Hz is what the wikipedia article says. I also thought it must have
been 10 kHz, One uses wikipedia at one's own risk. Gotta double check
things.

> > Not sure what that means. For this kind of sync, it is not
> > the frequency of the update, but the accuracy of it when it arrives.
> > This is a bit beyond my expertise.
>
> The 77.5kHz longwave carrier is also highly accurate and can be used to
> generate a very good PPS signal.

My question re the original post remains: do these devices use a
hardware interrupt line to do a pps?


Yours sincerely,

Roger Oberholtzer

OPQ Systems / Ramböll RST

Office: Int +46 10-615 60 20
Mobile: Int +46 70-815 1696
[hidden email]
________________________________________

Ramböll Sverige AB
Krukmakargatan 21
P.O. Box 17009
SE-104 62 Stockholm, Sweden
www.rambollrst.se


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
Roger Oberholtzer wrote:

> On Tue, 2012-08-28 at 16:05 +0200, Per Jessen wrote:
>  
>> > The DCF77 wikipedia entry (correctly?) states that inexpensive
>> > receivers typically have a 10 Hz bandwidth, and thus have this
>> > limitation.
>>
>> Did you leave out a 'k'?  I'm sure the bandwidth is more like 10kHz,
>> but I think that's only of important if you're using the carrier to
>> generate a PPS. See below.
>
> 10 Hz is what the wikipedia article says. I also thought it must have
> been 10 kHz, One uses wikipedia at one's own risk. Gotta double check
> things.
>
>> > Not sure what that means. For this kind of sync, it is not
>> > the frequency of the update, but the accuracy of it when it
>> > arrives. This is a bit beyond my expertise.
>>
>> The 77.5kHz longwave carrier is also highly accurate and can be used
>> to generate a very good PPS signal.
>
> My question re the original post remains: do these devices use a
> hardware interrupt line to do a pps?

Yes they do.  (well, I don't see how else it would be done with any
reasonable accuracy).


--
Per Jessen, Zürich (22.3°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

James Knott
Per Jessen wrote:
>> My question re the original post remains: do these devices use a
>> >hardware interrupt line to do a pps?
> Yes they do.  (well, I don't see how else it would be done with any
> reasonable accuracy).

Same as NTP.  Determine the average.  Over the short term, accuracy can
be very bad, but good when averaged over the long term.  That even
applies to electric wall clocks.  The power line frequency can drift
slightly in the short term, but long term it's very accurate.


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
James Knott wrote:

> Per Jessen wrote:
>>> My question re the original post remains: do these devices use a
>>> >hardware interrupt line to do a pps?
>>
>> Yes they do.  (well, I don't see how else it would be done with any
>> reasonable accuracy).
>
> Same as NTP.  Determine the average.  Over the short term, accuracy
> can be very bad, but good when averaged over the long term.

A PPS signal is always precise.  If you can't read it in a deterministic
way, the advantage of having it is gone.



--
Per Jessen, Zürich (21.3°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

James Knott
Per Jessen wrote:
> A PPS signal is always precise.  If you can't read it in a deterministic
> way, the advantage of having it is gone.
>

Are you familiar with phase locked loops?  They're often used to take an
unstable signal and make it stable.  The function includes LaPlace
transforms and such, but does a good job of creating a stable signal.  
Stability is a function of lock time, so if you want a accurate
interval, just give it plenty of time to stabilize.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Per Jessen
James Knott wrote:

> Per Jessen wrote:
>> A PPS signal is always precise.  If you can't read it in a
>> deterministic way, the advantage of having it is gone.
>>
>
> Are you familiar with phase locked loops?  They're often used to take
> an unstable signal and make it stable.

Yes, I know what PLLs are.

> The function includes LaPlace transforms and such, but does a good job
> of creating a stable signal. Stability is a function of lock time, so
> if you want a accurate interval, just give it plenty of time to
> stabilize.

Completely agree, but in a situation where you have a stable and very
precise signal (the PPS), if you can't interface with it properly,
there's no point in using it.


--
Per Jessen, Zürich (20.4°C)

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: DCF77, USB <-> parallel adapter, any solution?

Peter Maffter
In reply to this post by Per Jessen
> Von: Per Jessen <[hidden email]>

> An: [hidden email]
> CC:
> Gesendet: 14:39 Dienstag, 28.August 2012
> Betreff: Re: [opensuse] DCF77, USB <-> parallel adapter, any solution?

...
> The OP had a problem with a USB parallel port, my old fashioned serial
> setup is working fine :-) 

Actually it is a problem of hardware, drivers:
the hardware seems to be geared to usb <-> parallel solutions for printer
and therefore also the drivers seem to be missing the parts which are needed
for the DCF77 Conrad Parallel clock.

E.g. the DeLock adapter makes the following entry in the logfiles when attaching to USB:
Aug 29 00:59:39 mymachine kernel: [13805.322787] usb 2-1.4.7: new full speed USB device number 6 using ehci_hcd
Aug 29 00:59:39 mymachine kernel: [13805.408698] usb 2-1.4.7: New USB device found, idVendor=0fe6, idProduct=811e
Aug 29 00:59:39 mymachine kernel: [13805.408703] usb 2-1.4.7: New USB device strings: Mfr=0, Product=2, SerialNumber=0
Aug 29 00:59:39 mymachine kernel: [13805.408706] usb 2-1.4.7: Product: IEEE-1284 Controller
Aug 29 00:59:39 mymachine mtp-probe: checking bus 2, device 6: "/sys/devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.4/2-1.4.7"
Aug 29 00:59:40 mymachine mtp-probe: bus: 2, device: 6 was not an MTP device
Aug 29 00:59:40 mymachine udev-configure-printer: add /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.4/2-1.4.7/2-1.4.7:1.0
Aug 29 00:59:40 mymachine udev-configure-printer: device devpath is /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.4/2-1.4.7
Aug 29 00:59:40 mymachine udev-configure-printer: Device vendor/product is 0FE6:811E
Aug 29 00:59:40 mymachine kernel: [13805.497056] usbcore: registered new interface driver usblp

But I do not know how I get the
/dev/pcfclock0
/dev/pcfclock1
/dev/pcfclock2
working together with this.
mknod -m 444 /dev/pcfclock0 c 181 0
mknod -m 444 /dev/pcfclock1 c 181 1
mknod -m 444 /dev/pcfclock2 c 181 2
was the way when using it with the older parallel plugin and the pcfclock program
(which is currently pcfclock-0.44-250.1.2.x86_64.rpm).

One suggested NTP - I know that already. ;-)
But I want the former solution to work somehow.
Other ways: more expensive hardware like e.g.:
http://www.lindy.de/netzwerk-timeserver/20988.html

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

12