[GreenKeys] M28 using 20ma

[Harold Hallikainen]

>
> The 'pulling' magnets required the magnetic
> field from the loop current to physically
> attract the little plate against the selector
> magnet ...  this took a lot of 'oomph' as well
> as a few milliseconds, and also required the
> loop current to KEEP the armature pulled
> against the selector magnet.
>
> The 'holding' magnets have a cam that pushes
> the armature up against the selector magnet,
> doing mechanically what the pulling magnet
> setup had to do with loop current ... once the
> armature is pushed against the selector magnet,
> it was a lot easier to 'hold' it there with a lower
> loop current, such as the 20ma.
>
> otoh, as has been mentioned in some other responses, there
> is also the LR time constant [actually L/R] consideration
> since that affects the SHAPE of the square wave for each
> data bit.  The selector magnet, being a magnetic device,
> requires TIME for the magnetic field to be formed ...  how
> fast this happens is dependent on the L
> (inductance/sel.magnet) and the R (current limiting
> resistance in the loop).  That is also a reason why the
> 120vDC was used rather that some lower loop voltage
> like 12v ...  going back to the L/R time constant formula,
> you can't change the L of the selector magnet, so you
> want a larger R for faster response.
>
>   http://www.learnabout-electronics.org/ac_theory/dc_ccts45.php
>
> So for a fixed voltage (120vDC) and a fixed L (selector
> magnet) the controlling variable is the R ... so the higher
> the R, the faster the selector magnet pulls in.  Who cares?
> Because this affects the RANGE on the selector range
> finder on the typing unit.  A perfect square wave in the loop
> gives you a full range on the dial; a rounded square wave
> (due to time it takes to build up the magnetic field in the
> selector magnet) reduces your range, more and more toward
> the center of the rangefinder scale.
>
> So why not raise the R in the loop even higher and use the
> lower 20mA loop current?   Ah, because it is the CURRENT
> that causes the magnetic field in the selector magnet to
> build up!!   and that's why, for 20mA loops, they put the
> coils of the selector magnet in series to create more
> magnetic pull by adding the two coils; for 60mA operation of
> that type of selector magnet, the coils are in parallel so
> they don't get burned out from the high current ...
>

Great explanation! In addition, inductors (the selector magnets) in series
causes the inductance to add, while the inductance divides by two when the
magnets are in parallel (ignoring, for the moment, mutual inductance).
Considering mutual inductance, we could consider the selector magnets to
be one inductor with twice as many turns. This causes the inductance to go
up four times. Looking at the parallel magnets, the inductance drops to
half if there is no mutual inductance between them. If there is full
coupling, the inductance of the two in parallel is the same as one - It's
like we just wound the magnets with thicker wire. Also, the inductance
varies as the armature moves towards the selector magnet core. Then
there's core saturation causing the inductance to drop. So, a lot of
variables! Anyway, I always thought holding magnets were a very clever
idea. We are only interested in what the loop current is at the middle of
the bit time, so hold the armature until just before then, then see if it
falls away.

Harold



-- 
FCC Rules Updated Daily at http://www.hallikainen.com
Not sent from an iPhone.