<?xml version="1.0" standalone="yes"?>
<?xml-stylesheet type="text/xsl" href="..\..\XSLTemplates\CommandsStyleSheet.xsl"?>
<Command>
<Name>drum tick</Name>
<Description>
<DescriptionText><text>
measures the tick width by spinning the drum at (4 rpm) (8 rpm for VLF)
and timing the passage of each tick. A listing of the location, and % error
and % relative error is generated for the first 100 ticks outside of the error threshold selected.
</text></DescriptionText>
<DescriptionText><text>
&lt;threshold&gt; is a number in % of the ideal tick width.  If too many errors occur then the buffer
will overflow and the command will quit or auto restart with a higher threshold.
Use a higher threshold and only the errors that are more than the threshold will be
printed.
</text></DescriptionText>
<DescriptionText><text>
Typically, on PS3244, they are all within 6% for a clean disc and &gt;15% at dirty
spots on the disc.
</text></DescriptionText>
<DescriptionText><text>
On TS news (Heidenhain encoder) you can get maximum error of &lt; 5% - 7% tick width error.
</text></DescriptionText>
<DescriptionText><text>
VLF drums must be fairly well balanced for this command to work. Turn off the power
and move the balance weights and/or TEC till the drum is balanced. A good starting
point is TEC's parked and balance weights both on the opposite side to the TEC/LEC.
</text></DescriptionText>
<DescriptionText><text>
Typically, on VLF, errors are &lt; 14%, but the absolute max permissable is 18% which is
probably a defect in the disc.
</text></DescriptionText>
<DescriptionText><text>
Separate pid parameters are used for the slow speed spin, these must be optomised or
else the error discrepancies will excede 10% even for a perfect and clean disc because it is actually
the drum speed which is changing not the tick width which is in error.
</text></DescriptionText>
<DescriptionText><text>
Relative error is the % deviation of a tick width relative to the preceding 100 ticks.  This
reduces any effects caused by jitter in the drum velocity.
</text></DescriptionText>
<DescriptionText><text>
TICK &lt;INDEX&gt;
</text></DescriptionText>
<DescriptionText><text>
"Tick index" will spin the drum at 4 rpm and check the index pulse width and very carefully
check that the counter does not over count or under count just before the index.
</text></DescriptionText>
<DescriptionText><text>
TICK &lt; SA | SB | SI | PHASE &gt;
</text></DescriptionText>
<DescriptionText><text>
These will all spin the drum at 20 rpm and continuously print out data.  SA and SB will
print the SA and SB duty cycle in % (should be 50%), SI will print the deviation of
the SI pulse width from ideal in %, (should be 0% +/- 10% is the absolute limit),
PHASE will print out the phase angle in degrees between TachA and TachB (should be 90
degrees. This is set on Creo encoders by moving the position of the encode block and
requires an oscilloscope.)
</text></DescriptionText>
</Description>
<AppliesTo>
<UnnumberedList>
<ListItem><TextApplies>Platesetter_3244</TextApplies></ListItem>
<ListItem><TextApplies>Platesetter_VLF</TextApplies></ListItem>
<ListItem><TextApplies>Trendsetter_VLF</TextApplies></ListItem>
<ListItem><TextApplies>Trendsetter_400/800</TextApplies></ListItem>
<ListItem><TextApplies>Thermoflex</TextApplies></ListItem>
<ListItem><TextApplies>Trendsetter_400/800_Auto</TextApplies></ListItem>
</UnnumberedList>
</AppliesTo>
<Syntax>
<NumberedList>
<ListItem><Text>
drum tick &lt;threshold%&gt;
</Text></ListItem>
<ListItem><Text>
drum tick SA
</Text></ListItem>
<ListItem><Text>
drum tick SB
</Text></ListItem>
<ListItem><Text>
drum tick SI
</Text></ListItem>
<ListItem><Text>
drum tick phase
</Text></ListItem>
<ListItem><Text>
drum tick index
</Text></ListItem>
</NumberedList>
</Syntax>
<Notes>
<DescriptionText><text>
To find a dirt spot on the disc, run "drum tick x" starting at 6 for 3244 and 18 for VLF,
Repeat the command with different x until you get 5 - 20 errors. Pick the worst one and
type "drum move xxxxx h"  where xxxxx is the position of the largest error. Mark the drum
(a piece of masking tape works well) exactly in line with the encoder laser (which is
approx in line with the red led on the encoder).
Idle the drum (flyoff saves you walking), then clean the disc in the general area of your
mark on the drum.  Repeat for every position where the error is greater than 8% (3244) or
14% (VLF - this is not ideal but passable)
</text></DescriptionText>
<DescriptionText><text>
On PS3244 with a really clean disc, you can get a maximum absolute error of  &lt;5%
tick width error and relative error of &lt;3%.
</text></DescriptionText>
<DescriptionText><text>
On TS news (Heidenhain encoder) you can get maximum error of &lt; 4% - 6% absolute
tick width error.
( This error is probably mainly caused by the drum control PID which has to be perfectly
steady to get accurate results).
</text></DescriptionText>
<DescriptionText><text>
On a '"not quite clean enough'"  VLF disc the errors were &gt;25%
On VLF, the encoder discs are still quite bad.
Using the same VLF (VE448) that had been cleaned,  the best I could get was 16%,  and there
turned out to be a defect in the disc at every single place where the error was &gt; 11% (The
ones at 11% were very small, the ones at 18%  were quite easily visible
they measured 180um)
</text></DescriptionText>
</Notes>
<SeeAlso>
<UnnumberedList>
<li><Link><LinkAddress>../commands/drum.xml</LinkAddress><LinkTitle>Command: drum</LinkTitle></Link></li>
</UnnumberedList>
</SeeAlso>
<OtherLinks>
<Link><LinkAddress>..\..\ShowAllErrors.htm</LinkAddress><LinkTitle>Firmware 0 Error Index</LinkTitle></Link>
<Link><LinkAddress>..\..\ShowAllCommands.htm</LinkAddress><LinkTitle>Firmware 0 Command Index</LinkTitle></Link>
<Link><LinkAddress>..\..\ShowAllNVS.htm</LinkAddress><LinkTitle>Firmware 0 NVS Index</LinkTitle></Link>
</OtherLinks>
<Updated>
Updated: Wed Jul 16 19:38:58 2008 PST
</Updated>
<Links>
</Links>
</Command>
