Re: G201 death/repair/questions/problems (long post)

Hi,

This is a long post. For trobleshooting non-supported drives, see
immediately below. To set the record straight about these drives, see
the "P.S" at the end of this post.

Trouble-shooting non-supported drives:

1) Inspect the PCB for charring. If a fuse has not been used, enough
current can flow thru the drive after a fault to char (burn) the
board. If any burned or carbonized areas are noted, the drive cannot
be repaired. Carbon conducts electricity.

2) If it passes test (1) then test the power MOSFETS. Set a
multimeter to "Ohms" and measure each of the 4 phase output
resistance to ground terminal and then to the power supply terminal.
This requires 8 measurements, (1 to 3, 1 to 4, 1 to 5, 1 to 6, 2 to
3, 2 to 4, 2 to 5 and 2 to 6).

Good readings will be megohms, bad readings are less than 100 ohms,
usually a dead short.

3) If it passes test (2) then connect ONLY the power supply to the
drive, nothing else. Measure the DC power supply current flowing into
the drive. A good reading will be less than 0.03A, a bad reading will
be 100mA or more.

4) If it passes test (3) then measure the DC voltage across the
CURRENT SET terminals. A good reading will be 17.5VDC +/- 0.5V. Bad
readings will be above or below those limits.

What to do if it fails a test:
------------------------------
Fails test (1): Cannot be repaired.
-----------------------------------
Fails test (2): Remove the 4-40 cap-head screws and remove the board.
Turn it upside down. The MOSFET pair closest to the main connector go
to PHASE D, the next pair to PHASE C and so on.

Identify the dead MOSFET / MOSFETs and clip the 3 leads going to the
part. Use a pair of tweezers and a medium soldering iron with a wedge
tip and carefully remove each remaining lead from the PCB. Use solder-
wick to clear the 3 holes of solder.

Apply ONLY the power supply (24VDC) and a STEP pulse source to the
drive. Measure the power supply DC amps going to the drive. A good
reading will be 0.03A or less and makes the drive a good candidate
for repair.

(1) If so, use a non-hydroscopic solvent like xylene (do not use
acetone) on a Q-tip to clean the board of rosin flux residue. Take
the replacement IRF530 and bend its leads back at the natural bend
location 45 degrees past 90 degrees. Press-fit the MOSFET into place
taking care its body seats directly onto the PCB.

(2) Re-apply power to the drive and measure the current draw. If it
stays below 0.03A then the drive may be repaired. Solder the MOSFET
into place AFTER the board has been re-seated with the 4-40 screws.
Clip the excess lead lenght and re-try the drive with a motor.

(3) If it does not work, the damage has propogated past the dead
MOSFET and has killed the IR2104S half-bridge driver. This requires
SMT rework equipment (hot air soldering).
---------------
Fails test (3):
---------------
This means one of about 50 components connected across the internal
12VDC supply has shorted.

(1) You need a variable voltage, variable current limit lab-type
power supply. Set the supply to +12VDC and turn the current limit to
zero. Clip the "+" supply lead to the opto pin 8 and the gnd lead to
the opto pin 5.

(2) Slowly turn up the current limit until at least 6VDC is indicated
on the supply. Use your fingertip to search out a part on the board
that will be hot, usually an IC. If it is hot, replace it and try the
test again. When you see only 0.03A or less, the drive may be
repaired.

-------------
Fails test 4:
-------------

If the voltage is below limit, go to "Fails test 3:" above. If it is
above the limit then the main TL783C voltage regulator has failed and
the drive is probably unrepairable due to multiple/all ICs being
destroyed.

Mariss

P.S. About these drives:

On our very first production run of the G201s the sheetmetal vendor
supplied soft anodized plates instead of the specified hard
anodizing. Our options were to rework the first revision drives or
sell the the second revision (auto standby) with the proper plates.
We opted for door #2 because it was quicker; we had to meet orders.

The choice now was to junk the first 1,000 drives or to rework them
as time allowed.

Again we chose door #2; all of the first revision drives had their
plates replaced with the hard anodized ones but now they were one
generation behind and thus rendered unsaleable at the list price.

Because they were one generation behind, we sold them off to
resellers at cost ($49) and licked our wounds; hundreds of hours of
hard work with nothing to show for it.

Main point: No drives were ever sold with the soft plates; I would
never sell anything I knew was junk or flawed in any way. I have a
large box filled with 1,000 plates to remind me we did the right
thing.


--- In CAD_CAM_EDM_DRO@yahoogroups.com, Steven Ciciora
<sciciora@A...> wrote:

> I thought someone told me that the Geckos that were sold without

lids

> initially had the heat sink/base plate 'soft' anodized, not 'hard'
> anodized. Small scratches in the heat sink/base plate would cause

the case

> of the FETS to conduct/short out to the base plate. This is a bad

thing.