Ericsson chips, Camtronics driver

On 18 Jan, CAD_CAM_EDM_DRO@onelist.com wrote:

> Message: 4
> Date: Mon, 17 Jan 2000 10:41:46 -0700
> From: Roger Brower <rbrower@...>
> Subject: Ericsson chips, Camtronics driver


> I'm having trouble getting mine going,
> so I presume I've fried some parts. If
> I'm going to be able to count on this
> driver, I'll need spares. One axis used
> to work well, but I've had to replace
> the 7805 twice, so I added a heat sink
> and 1A fuse in +5v circuit. How much
> should it really be drawing?

IIRC, about 120 ma per channel. The 7805 is not stressed
in this design.

> The IRF9Z24 and IRFZ34 Mosfets are about
> .89 ea in Digikey P232. Does anyone know
> how to test them in circuit? I have 5A
> breakers in the three +36V lines, but I
> haven't shunted them through the motor
> connectors yet; Does it really blow them
> out if no motor is being driven?

The only ways to blow them up in circuit that I can
think of is

a damaged 3776.
solder bridges
excessive supply voltage
reversed supply voltage
incorrect insertion.

Double check that you've got them in the right way round, solder
bridges, all the normal stuff.

Also, check your motor supply voltage. The absolute max for the 3776
is 45 volts. I'm running mine at 42, but I have my fingers crossed
as well.


A cheap and nasty test is continuity with a low current
ohmmeter ( a typical digital meter will do fine ) between
gate and source and drain. Bad ones will show as almost
shorts.

Be aware that these ARE static sensitive devices - the
manufacturers aren't kidding about this. You CAN blow them
up by touching them, and depending upon the part of the
country you're in, you WILL blow them up.

If there's one consolation about living in the wet Pacific
Northwest, it's that it rains so much that static is rarely
a problem. However, even here, on a clear winter day, static
blows things up.


Get rid of your breakers ! Waste of wire to connect them up.
A typical breaker takes a 100% overload for 3-5 seconds to trip, with
some being much worse than that. Given that the FETs will fail in
nanoseconds, a breaker is not going to protect much. Breakers protect
the wire, not the electronics.

A fuse won't give much better protection either. I remember
seeing a spec sheet for a fuse manufacturer, where "time to
open" was compared to overload.

A 1 amp fuse with 1.1 amps going through it was "guaranteed" to
open within THIRTY seconds ! Even at 10 amps, the time was some
large number of milliseconds ( 100 ? ) to open.

Fuses WILL protect the MOTORS from a particular form of electronics
failure and given that the motors are probably the most expensive thing
in the system, a fuse should be included. The SIZE of fuse might be a
surprise, though, as the correct size is related to the characteristics
of the fuse, of course, the motor current AND the supply voltage. The
higher the supply voltage, the lower the current rating of the fuse !

For a typical 5 amp system with a 36 volt supply, I'd start with 1 amp
fuses. Bet they never blow.

> I'm intending to switch the main fan
> through a temp sensor once I'm running
> any big amps, and will implement the
> mod's to shield the signal lines and
> replace the 0R1 Ohmites once I'm driving
> through the Parallel port. What kind of
> non-inductive resistors am I looking
> for?

You may want to save yourself some trouble and just wire the fan on and
leave it on. If fan noise is going to bother you over the noise of your
machine, you need better hearing protectors.


You'll have to look hard through the Digikey catalog, but they DO have
some non-inductive resistors in there. Frankly, hunting them down
would not be at the top of my list, even though all the documents tell
you to use them. The inductance of a .1 ohm resistor is very low, and
with the RC filter on the input to the comparator of the 3776, most
ringing is damped out.

I looked hard at the circuit with my fancy scope with both types of
resistors and I couldn't see any great difference between them. It
could be that the characteristics of my motors were simply swamping out
any effects of the resistors, but even so, I don't think you should
lose sleep over the issue.


On the subject of shielding, if shielding is required for clean
operation, you may have a ground problem, so double check your ground
wiring. Also, you may want to consider using optoisolators on the
signals, as they will isolate the grounds, of course, but also give a
lower impedence source that is less troubled by rf noise.

> Thanks in advance- it's great to have
> this forum available, as I'm new to
> electronics and quite on my own in this
> little one horse town...

Great group of people on this list, aren't there ?

Alan
--

Alan Rothenbush | The Spartans do not ask the number of the
Academic Computing Services | enemy, only where they are.
Simon Fraser University |
Burnaby, B.C., Canada | Agix of Sparta