My comments are interspersed with yours below:
K2XT: "Nice work Jerry. I don't completely understand all, but it
sounds to me as if you are saying something like the lines (i.e. pins
1-2, and 3-4 going to the driven element) are balanced. Meaning no
current is returning on ground?".....
K4SAV: Yes, no current on ground. Current up and back to the motors are
only on the two wire pair. The power supply is floating relative to
case ground. However if you use the cable which interfaces to the
radio, the negative side of the power supply then becomes grounded.
That is still OK with respect to signals to the motors, still no signal
current on ground. The wires to the motor are only connected to the
controller, not to ground at any place on the antenna.
K2XT: "To provide some means of lightning protection on the control
lines we need to consider two scenarios. One is to protect the
controller, specifically the driver chips. The other is to "keep the
fireball out of the house", meaning keeping surges from getting to the
controller where they can work their way to other devices such as a
transceiver connected to the data port. So, in the latter case, it
seems to me MOVs across all conductors to ground, placed on a SPG at the
house entrance, would serve that purpose the same as a rotor protector
using MOVs, or gas tubes on the coax lines. "...
K4SAV: As far as keeping the fireball out of the house that should be
done with proper grounding at the tower. Anything that remains on lines
to the shack should be small transients. Placing MOVs on those lines
may reduce the amount of transient energy going into the controller and
into other devices like the radio and computer. HOWEVER if a transient
is large enough to trigger the MOV, the amount of reduction offered by
the MOV is not enough to protect the controller. Depending on what
fails inside the controller will determine if the radio or computer also
fail. The MOVs may reduce the possibility of radio and computer failure
because of lower transient voltages going onto the controller. However
if a failed controller puts plus or minus 39 volts on the line going to
a radio serial port I doubt the radio will survive undamaged. I'm not
willing to test this theory and I don't have the controller schematic to
research the answer.
K2XT: "The question is, do those MOVs protect the driver chips, due to
their bridge configuration?"....
K4SAV: No they don't. The MOVs will not limit the current into the
driver chips to 1 amp or less, which is the amount specified by the
driver IC manufacturer as the maximum allowed before damage. Notice
this statement is in direct contradiction with that being given out by
the people at SteppIR.
K2XT: "And if the MOVs should short out after doing their job, does the
controller then destroy itself due to seeing a direct short to ground on
one or more of the output lines?
K4SAV: Yes it does. If you plug the controller into a line that is
shorted by an MOV, the then driver chips will blow. I have personally
verified this. I used to have MOVs on the lines, but then realized they
were more of a liability than a help. I disconnect everything when I'm
not at home or storms are near. With MOVs on the lines, should one fail
while I am not at home, I might not suspect that one was shorted to a
lightning event. Then when I plug the controller back in, it will blow
due to the shorted MOV. I removed the MOVs from these lines.
Disconnecting everything is the only solution I have found to protecting
the driver chips.
K2XT: "What about MOVs across the motor windings (pins 1-2, 3-4, etc of
the DB25)? to protect the chips."....
K4SAV: If you mean putting the MOVs on at the antenna, that won't
protect the driver chips. It might help protect the stepper motors, but
those are pretty robust and I haven't heard of anyone loosing a motor
due to lightning. If you mean putting the MOVs across the wire pairs
(and not to ground) that won't help either. The normal voltage
excursions on these lines relative to ground is between a low voltage
and something close to the power supply level (a little less). Looking
at the two lines differentially the voltage excursions are approximately
plus and minus the supply voltage (a little less). So the MOVs would
have to limit the current, not limit the voltage. MOVs don't work like
that.
K2XT: "Incidentally, MOVs rated at 39 volts are available for 45 cents
at Mouser. They are rated for huge currents, 2000 amps if memory serves
me correctly. "....
K4SAV: Yes you can buy MOVs of all sizes. You have to pay attention to
the capacitance of the large ones when used on circuits that switch.
Jerry, K4SAV
Rick Stealey wrote:
>
>
>> Providing protection for the driver requires a circuit that limits the
>> output current on every line
>>
>
> Nice work Jerry. I don't completely understand all, but it sounds to me as
> if you are saying something like the lines (i.e. pins 1-2, and 3-4 going to
> the driven element)
> are balanced. Meaning no current is returning on ground?
> To provide some means of lightning protection on the control lines we need to
> consider two scenarios. One is to protect the
> controller, specifically the driver chips. The other is to "keep the
> fireball out of the house", meaning keeping surges from getting to the
> controller where they can work their way to other devices such as a
> transceiver connected to the data port.
> So, in the latter case, it seems to me MOVs across all conductors to ground,
> placed on a SPG at the house entrance, would serve that purpose the same as a
> rotor protector using MOVs, or gas tubes on the coax lines.
> The question is, do those MOVs protect the driver chips, due to their bridge
> configuration? And if the MOVs should short out after doing their job, does
> the controller then destroy itself due to seeing a direct short to ground on
> one or more of the output lines?
> What about MOVs across the motor windings (pins 1-2, 3-4, etc of the DB25)?
> to protect the chips.
> Incidentally, MOVs rated at 39 volts are available for 45 cents at Mouser.
> They are rated for huge currents, 2000 amps if memory serves me correctly.
> Rick K2XT
>
>
> _
>
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|