[ARC5] Drift in BC-453 - more

Tue Dec 12 20:02:39 EST 2017

Hi

A constant ppm drift is a pretty good first approximation to what is likely to be happening. You can compensate 
this or that. It works about as well at this frequency or that frequency provided you don’t get too crazy. The net 
result is some drift in ppm. Might you be off 2:1 or 3:1 with that assumption over a wide range? Sure, you might. 
You still are in the general ballpark when trying to work out what to expect. That’s fairly important when trying 
to work out “is it broke?”. You don’t *always* need to have a duplicate device to compare to. Troubleshooting is 
as much about “does that make sense?” as anything else. 

Bob

> On Dec 12, 2017, at 7:30 PM, Bill Cromwell <wrcromwell at gmail.com> wrote:
> 
> Hi,
> 
> If the drift in parts per million is the same for the oscillator systems in each of those radios is the same then all of that holds up. As soon as we do anything to compensate for drift your statements fall apart. But generally free running at lower frequencies are more stable than higher frequency oscillator systems. Temperature is the biggest bogey man in the drift arena and there are well known ways to manage that. Hams and professional engineers alike have built some pretty stable oscillator systems (heterodyne HFO, BFO, VFO combinations).
> 
> So we can build a radio to receive WSPR and it's total drift stack comes to 1.9 Hz over two minutes. What if the distant TX also drifts a mere 2.9 Hz during the same two minutes but drifts the other direction. Then add the same kind of atmospheric mayhem that causes WWV to be "useless". Again, we do the best we can.
> 
> I think it was mentioned again that time has to be within a second or two at each station. The use of network time standards is indicated. My (mis)understanding is that the "esp" modes rely on online databases to help sort out what is being sent - to improve the chances of a correct guess. So those milliHertz modes seem to depend on outside resources to prop up their useability. All of that for tiny a message that takes all night to get there. Sort of like "Kilroy was here".
> 
> That is not my cup of tea at all. This is not to dismiss WSPR or JT(favorite flavor here). Splitting milliHertz seems like a reasonable facet of the hobby to me. Other hams seem to be happy with it. We all have to control our transmissions. The new MF and LF allocations are very narrow slices of spectrum and we need to be aware of what our signals are doing and how they can be fitted together in that narrow slice. So some degree of splitting hairs is required. I believe (perhaps incorrectly) that I can see (not hear) a difference of less than one cycle per second. Maybe in my lab it is only deciHertz and not milliHertz or microHertz. That will have to be good enough.
> 
> As to the original post..the OP seems to be experiencing a wildly excessive amount of drift. I run my command receivers on batteries and when I pipe the audio into a sound card and display it on a waterfall display I cannot see *any* drift. That means no net drift for the VFO, BFO, incoming signals, and the computer clocks all combined. Maybe there is a deciHertz that I am unable to resolve. In *my* radio world that is insignificant. Does the OP have truck tire tracks across his radio? <evil grin>
> 
> 73,
> 
> Bill  KU8H
> 
> On 12/12/2017 03:26 PM, Bob kb8tq wrote:
>> Hi
>> 
>> If you are running a radio at 400 KHz, and a similar radio at 4,000 KHz the drift of the
>> second one (in Hz) probably will be 10X the drift of the first radio. The same applies
>> if you go to 40 MHz (you now are at 100X vs 400 KHz). Yes, there are a bunch of
>> assumptions there about radio construction, IF frequencies and various other things.
>> 
>> To the extent this is “simple”, a 1 Hz drift on a BC-453 would be about the same as a 100 Hz
>> drift on a similar radio on 10 meters or a 10 Hz drift on 80 meters. Simply put - if you *think*
>> you could listen to SSB on 10M with a low IF / single conversion non-crystal based
>> radio (ignoring anything but drift as a problem) … you should get about 1 Hz drift on the 453.
>> You could compare to comfortably listening to SSB on 80 M as well.
>> 
>> Of course, you immediately get into “drift over what period of time”. Any radio like this that I
>> ever used for SSB stayed on pretty much all the time. Use it in the evening and leave it on
>> all day while I was busy with other stuff. Starting from dead cold … yikes …. that’s a tough
>> test for a tube based radio.
>> 
>> If you *do* want to test for sub 1Hz sort of stability it’s likely easier to shove a stable signal
>> into the radio and let a computer look at the audio coming out than to do a bunch of other
>> crazy stuff. The advantage is that the result of the test is looking at the same thing the
>> decoder software in the computer will be looking at.  There are a lot reasonably stable
>> oscillators on the auction sites to provide an adequate signal. The risk is more spending
>> $20 and getting a dead one than anything else. Pretty much anything called an OCXO
>> should do the trick.
>> 
>> Lots of fun
>> 
>> Bob
>> 
>>> On Dec 12, 2017, at 2:57 PM, Kenneth G. Gordon <kgordon2006 at frontier.com> wrote:
>>> 
>>> On 12 Dec 2017 at 14:32, MICHAEL ST ANGELO wrote:
>>> 
>>>> Bill,
>>>> 
>>>> What is this millihertz spec for WSPR? I haven't tried it yet but I
>>>> thought frequexcy stavility has to be a couple of hertz for the 2
>>>> minute cycle:
>>>> 
>>>> <https://groups.google.com/forum/#!topic/2-meter-wspr/-hDK071K1iM>
>>>> 
>>>> Mike N2MS
>>> 
>>> Hi again, Mike. That spec is for 2 meters: it becomes far, far more critical at HF and MF.
>>> Yes. Millhertz.
>>> 
>>> Since, "Many WSJT-X capabilities depend on signal-detection bandwidths no more than a
>>> few Hz. Frequency accuracy and stability are therefore unusually important."
>>> 
>>> See this for details:
>>> 
>>> http://www.physics.princeton.edu/pulsar/K1JT/wsjtx-doc/wsjtx-main-1.8.0.html#GENERAL
>>> 
>>> In fact, someone on one of our lists mentioned something like <0.5 Hz drift in 2 minutes or
>>> less. I can't remember the details now, but his post included the link above. As I remember
>>> it, he also told us that if the drift exceeded what I mentioned above, that would prevent
>>> decoding of the input.
>>> 
>>> I know that Bill Cromwell noted enough drift from the very slight drift in filament voltage by
>>> line-voltage drift to cause very noticable drift. As he says, he decided to simply forego
>>> attempts to use those modes.
>>> 
>>> Although I am inclined to agree with him, I am very curious to learn if it IS possible to use a
>>> BC-453 for these modes. I would guess not, but stranger things have happened.
>>> 
>>> Ken W7EKB
>>> 
>>> ---
>>> This email has been checked for viruses by Avast antivirus software.
>>> https://www.avast.com/antivirus
>>> 
>>> ______________________________________________________________
>>> ARC5 mailing list
>>> Home: http://mailman.qth.net/mailman/listinfo/arc5
>>> Help: http://mailman.qth.net/mmfaq.htm
>>> Post: mailto:ARC5 at mailman.qth.net
>>> 
>>> This list hosted by: http://www.qsl.net
>>> Please help support this email list: http://www.qsl.net/donate.html
>> 
>> ______________________________________________________________
>> ARC5 mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/arc5
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:ARC5 at mailman.qth.net
>> 
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> 
> 
> -- 
> bark less - wag more
> ______________________________________________________________
> ARC5 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/arc5
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:ARC5 at mailman.qth.net
> 
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html