[GreenKeys] 1 or 1.5 stop bits?

Mon Jan 3 08:46:54 EST 2022

And it gets even more interesting... After the 5th bit is sampled, the print stroke can begin. The print stroke takes about 5 bit times, giving a total of about 12 bit times per character (given one bit stop time at the end to be idle for the next character). This timing gives only 38 wpm - which is obviously slower than what really happens.

To get around this problem, the mechanism overlaps printing of the current character with reception of the next character. To achieve this, the selector portion of the shaft rotates independently of the printing section of the shaft. The print cam section isn’t released to turn its revolution until after the 5th data bit is sampled. Meanwhile, while the print section is performing the print stroke, the selector section can rotate again and receive the next character. If you hand crank a machine running open loop, you’ll see the print stroke finishes about the time data bit 4 of the next character is being sampled.

During the print stroke, the selector vanes are locked into position for the character being printed, so how can the selector decode the bits of the next character if the selector vanes are locked in place for the current character? The selector mechanism actually “stores” the new bit values via the selector lever spring for each bit. The spring is what pulls the selector sword forward into the T-lever that then tilts the selector vane. While the vane is locked in place, the spring can’t pull the sword forward because the T-lever won’t move. As soon as the print stroke completes and the selector vanes are released, the selector springs can move the vanes for the already received bits. If you hand crank through characters with opposing bit patterns, you can see the first 4 selector vanes all change at the same time when the previous print stroke completes.

Mike

> On Jan 2, 2022, at 10:54 PM, Harold Hallikainen <harold at w6iwi.org> wrote:
> 
> OK, 143ms per rotation would make the shaft stop 11ms after the end of the
> last data bit. Or, it stops in the middle of a single stop bit.
> 
> I just looked at the distributor on my model 14 TD. I measured the chord
> length of the stop bit and found it to be 16.2mm. The chord length of a
> data bit measured 11.4ms. Recognizing that chord length is only an
> approximation of the arc length, we see that the stop bit is about 1.42
> times the length of a data bit. Using the typical 31ms stop and 22ms data,
> we get a ratio of 1.41. Pretty close!
> 
> Also, when the distributor stops, it stops in the middle of the stop bit.
> So, when it starts rotating again (when a new tape is started), the idle
> stop bit continues for a while before the start bit starts. This allows
> the shaft to get up to speed before critical timing is required.
> 
> These are amazing machines!
> 
> Harold
> https://w6iwi.org
> 
> 
> 
>> On Sun, January 2, 2022 9:07 pm, John W9DDD wrote:
>> The receive shaft in a 45.45 baud printer the shaft revolves at 420 RPM.
>> That figures out to be 142.85714 mS per rotation. Of course there is some
>> variation, since depending on type of clutch, there's some delay getting
>> things moving.
>> 
>> John, W9DDD
>> 
>> 
>>>> On Jan 2, 2022, at 9:41 PM, Harold Hallikainen <harold at w6iwi.org> wrote:
>>> 
>>> When does the receive shaft stop rotating (on, for example, a model 15).
>>> I'd expect the sequence to be something like this:
>>> 
>>> 0 ms - start bit starts and shaft starts rotating
>>> +22 ms - start bit stops and bit 0 starts
>>> +11 ms - Middle of bit 0, bit 0 sampled
>>> +11 ms - end of bit 0, start of bit 1
>>> +11 ms - Middle of bit 1, bit 1 sampled
>>> +11 ms - End of bit 1, start of bit 2
>>> +11 ms - Middle of bit 2, bit 2 sampled
>>> +11 ms - End of bit 2, start of bit 3
>>> +11 ms - Middle of bit 3, bit 3 sampled
>>> +11 ms - End of bit 3, start of bit 4
>>> +11 ms - Middle of bit 4, bit 4 sampled
>>> +11 ms - End of bit 4, start of stop bit
>>> +11 ms - Possibly sample stop bit here and stop shaft rotation
>>> +11 ms - End of stop bit if it were 22 ms
>>> +9 ms  - End of stop bit with length of 31ms, start of start bit of next
>>> character
>>> 
>>> So, does the receive shaft stop rotating 11 ms after the end of of bit
>>> 4?
>>> It seems like this would be a good place to allow for speed variation
>>> between transmitter and receiver. BUT, perhaps it is sampled later,
>>> hence
>>> the longer stop bit.
>>> 
>>> With 22ms bits and 22ms stop, we'd set the UART to 5N1 (5 data bits, no
>>> parity, 1 stop bit).
>>> 
>>> With 22ms bits and a 31ms stop bit (which is what I learned a little
>>> over
>>> 50 years ago), it would be 5N1.4 (5 data bits, no parity, 1.3 stop
>>> bits).
>>> 
>>> So, does the receive shaft stop rotating 143 ms after the leading edge
>>> of
>>> the start bit (22 ms after the sampling of the most significant bit), or
>>> some other time?
>>> 
>>> Harold
>>> 
>>> ______________________________________________________________
>>> GreenKeys mailing list
>>> Home: http://mailman.qth.net/mailman/listinfo/greenkeys
>>> Help: http://mailman.qth.net/mmfaq.htm
>>> Post: mailto:GreenKeys at mailman.qth.net
>>> 
>>>>>> Jordan Spencer Cunningham's GreenKeys Search Tool:
>>>>>> https://teletype.net/gksearch
>>>>>> 2002-to-present greenkeys archive:
>>>>>> http://mailman.qth.net/pipermail/greenkeys/
>>>>>> 1998-to-2001 greenkeys archive:
>>>>>> http://mailman.qth.net/archive/greenkeys/greenkeys.html
>>>>>> Randy Guttery's 2001-to-2009 GreenKeys Search Tool:
>>>>>> http://comcents.com/tty/greenkeyssearch.html
>>> 
>>> This list hosted by: http://www.qsl.net
>>> Please help support this email list: http://www.qsl.net/donate.html
>>> Message delivered to w9ddd at tapr.org
>> 
>> 
> 
> 
> -- 
> https://w6iwi.org
> ______________________________________________________________
> GreenKeys mailing list
> Home: http://mailman.qth.net/mailman/listinfo/greenkeys
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:GreenKeys at mailman.qth.net
> 
>>>> Jordan Spencer Cunningham's GreenKeys Search Tool: https://teletype.net/gksearch
>>>> 2002-to-present greenkeys archive: http://mailman.qth.net/pipermail/greenkeys/
>>>> 1998-to-2001 greenkeys archive: http://mailman.qth.net/archive/greenkeys/greenkeys.html
>>>> Randy Guttery's 2001-to-2009 GreenKeys Search Tool: http://comcents.com/tty/greenkeyssearch.html
> 
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to deramp5113 at yahoo.com