It's true most micros won't run instructions that fast, but outputting a
count of pulses can often be handled by a timer peripheral at anything up
to the system clock rate. Alternatively a microprogrammable peripheral such
as the RP2040 PIO, Beagleboard PRUs, or the  venerable Motorola TPU.


On Thu, Jun 13, 2024 at 4:26 PM gregebert <[email protected]> wrote:

> I'm planning a similar approach with the HP5245L, but instead I will run
> the counter with a 10-second gate. The gate signal is an *output* from the
> counter, and while the signal is active, it just acts like a regular
> counter. So, all I need to do is wait for gate to go active, send a stream
> of pulses (up to 235,959 of them), and they will display the time.
>
> The odd thing about the 5245 is that the display is on while counting;
> annoying if you are measuring frequency but just what you need for a clock.
> After the initial count is displayed, which is pretty fast at 50Mhz (less
> than 5msec), the next 9+ seconds only require a single pulse to advance the
> seconds *in most cases*. When there is a rollover for the next minute,
> you need to send 41 pulses so that xx:x1:59 becomes xx:x:2:00 . Similar
> exceptions for ten-minute, hours, etc. At midnite, this trick wont work so
> you may as well flood the device with an erratic burst of pulses to make
> most of the digits roll.
>
> Since a 50Mhz pulse-output is too fast for a cheap microcontroller, it
> will be done from an FPGA.
>
> Of course I wont keep this thing running 24/7 because of the noise and
> wasted electricity. But at least I will get more usage out of th3 5245L.
>
> On Thursday, June 13, 2024 at 7:23:51 AM UTC-7 Adrian Godwin wrote:
>
>> Interesting idea.
>> I'd speculate that the accuracy problem where the counter failed to see
>> some pulses is perhaps a synchronization problem. If the pulse-counting
>> method worked by generating a variable number of pulses at 240kHz then they
>> should always be visible to the counter as long as that's within it's
>> range. But if the gate period of the counter didn't coincide with the
>> gating of the pulses at the generator, you would get variable results. And
>> there's no obvious way to make them synchronise without stop/start signals.
>>
>>
>> On Thu, Jun 13, 2024 at 3:04 PM Paul Andrews <[email protected]> wrote:
>>
>>> Why not both? Here is my partially successful attempt to display the
>>> time on a frequency counter by generating the correct frequency that would
>>> result in the time being shown in the frequency counter. I had forgotten
>>> about this until I read this thread:
>>>
>>>
>>> https://www.nixies.us/2018/03/25/display-the-time-on-an-old-frequency-counter/
>>>
>>> On Thursday, June 13, 2024 at 5:53:32 AM UTC-4 Adrian Godwin wrote:
>>>
>>>> I much prefer Nixies in functional equipment rather than for their own
>>>> sake. And while displaying them in a clock is fine, there are other types
>>>> of clock to cherish too, leading to a loss of space. And since I can build
>>>> a clock myself, other functionality is attractive (admittedly I wouldn't
>>>> pay $300).
>>>>
>>>> It's a pity there aren't a lot of devices other than counters and DVMs
>>>> to find - I already have Philps, Racal and Avo examples. I'd like to find a
>>>> nixie-based measurement system such as a Heidenheim.
>>>> Are there any other types of instruments that were made ?
>>>>
>>>>
>>>> On Thu, Jun 13, 2024 at 2:03 AM <[email protected]> wrote:
>>>>
>>>>> Re: [neonixie-l] A bit $teep for a DMM
>>>>> On Wed, Jun 12, 2024 at 11:57 AM Mac Doktor <[email protected]>
>>>>> wrote:
>>>>>
>>>>>> What makes this DMM so valuable? There are only four tubes to yank
>>>>>> out:
>>>>>>
>>>>>> https://www.ebay.com/itm/405017664588
>>>>>>
>>>>>>
>>>>>> It is a FLUKE, you would  only Yank the tubes to clan the pins and
>>>>>> remove the dust from the sockets... then reassemble carefully and bow 
>>>>>> down
>>>>>> in Veneration.
>>>>>>
>>>>> --
>>>>>
>>>> You received this message because you are subscribed to the Google
>>>>> Groups "neonixie-l" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>>> an email to [email protected].
>>>>>
>>>> To view this discussion on the web, visit
>>>>> https://groups.google.com/d/msgid/neonixie-l/194ac98d77051b3026a5ca51c003969b%40zeusprune.ca
>>>>> <https://groups.google.com/d/msgid/neonixie-l/194ac98d77051b3026a5ca51c003969b%40zeusprune.ca?utm_medium=email&utm_source=footer>
>>>>> .
>>>>>
>>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "neonixie-l" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to [email protected].
>>>
>> To view this discussion on the web, visit
>>> https://groups.google.com/d/msgid/neonixie-l/b082f74b-146c-48b7-aa06-f325bbceec92n%40googlegroups.com
>>> <https://groups.google.com/d/msgid/neonixie-l/b082f74b-146c-48b7-aa06-f325bbceec92n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>>
>> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/2fdbf720-532c-4590-b187-0fa16e79a999n%40googlegroups.com
> <https://groups.google.com/d/msgid/neonixie-l/2fdbf720-532c-4590-b187-0fa16e79a999n%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/CALiMYrsBw6MX7E5UMFbwh2rq1F_i0QuOnfudSFshAFyr-op-ZA%40mail.gmail.com.

Reply via email to