Re: Conversions

["J. Frederick Bartlett (bartlett@fnal.gov)" <bartlett@d0mino.fnal.gov>]

  I would like to add my support to Rolf's suggestion. For just the
reasons that he mentions, I have found the present algorithm for
specifying conversions to be awkward and confusing. Almost every time
I have to explain the scaling to one of my D0 colleagues (admittedly,
they are mostly physicists) I get this far-away, blank look.

							Fritz

> In the recent debate on conversions, someone mentioned doing new analog
> record types in a future release. Just to muddy the waters a bit more:
> I always disliked specifying linear conversions by giving ranges or
> EGUF,EGUL, because it forces you to come up with phantom numbers if the
> control voltage range of a device doesn't match the DAC/ADC ranges or if
> you have to deal, for example, with a 5 V DAC and a 10 V ADC.
> I had an old control control system, where we spec'd the conversions in
> two stages:
> - volts/bit and volt_offset to take care of the DAC/ADC
> - egu/volt and egu_offset to take care of the device.
> For efficiency, these two sets of values were combined into one
> calculation, but it was also possible to display the voltage values,
> which was quite helpful for the techs who probed the breakout panels
> with their DVMs.
> Just food for thought.
> - rolf -
> -- 
> +-------------------------------------------------+
> | Rolf Keitel, TRIUMF         Tel: (604) 222-7453 |
> | Group Leader ISAC Controls  Fax: (604) 222-7307 |
> | and Electronics Development                     |
> | Vancouver, B.C., Canada                         |
> +-------------------------------------------------+