[Andy Fritz]

I have a -94 4020 which cuts a very nice contour, but not very round circles.
I just searched for this and found that balancing the amps and then doing an "angle test" is needed. On the balancing, I have done this lots of times, but am not really sure what the following error should be. The manuals say 302 for DC motors and metric ball screws which is what I have. In the 05811 program it also says 302 for metric screws, but nothing about DC or AC motors. I have also seen somewhere that it depends on if it is a 700 ipm machine.

I just bought a new amp 0021 for the X and there was some guidlines with it which also said 302 for metric screws.

How do I find out if I have a 700 ipm machine or not? In the parameters I can select between 400 - 1000 ipm, but I assume the software is generic. right now it is set to 500 ipm, but I may have set it to this at some point as I had problems with servo errors on rapid moves a year ago.

Anyway for the time being I am using the 302 following error. It is never 302 solid during the balance program, but fairly close. I cant get it to zero when the program is stopped though. X is toggeling between 1 and 3 and Y is between 2 and 5. On a fluke meter the balance is spot on and exacly the same 0,735 VDC and -0,735 VDC on both X and Y. A Fadal tech once told me that using the Fluke meter for balancing should give better circles than adjusting for zero on stop, but I am getting eggs. Not circles =)

On the comp pot; the guidlines said to turn it CCW until motor vibrates and then got back Cw until it stops and then back off further a full turn. The full scale of the pot is not even one turn, so this could be either a typo and they mean "one marking" or it is not a typo but referring to another amp with several turns scale. Any advise on amp 0021?

Ok so how bad is it? Right now I am doing a part which has a 2 inch diam. boss and this is coming out as 2,008" measured in the X axis and 1,902" in Y axis so even if I compensate I would be 0,016" out of round.

I read in a post here that I should also do an angle test. It looked like this was another Fadal test program, but I dont have it and searching maintenance manual and diagnostics for "angle test" gives not hits.

Could someone please describe how to do it?

Another hint; The Y axis amp pulls around 3 amps on idle and the X pulls only 1,5. The A axis amp pulls 0,5 amp on idle and the Z around 2 amp. A guy from a Fadal service shop said that the backlash should be set a half the lost motion value as this would put less load on the amp. I tried this but it did not help. However if I increse the backlash comp setting I can drop the amp pull on idle from 3 to 1,5. Hmmm, sounds like the increased value is what it should be set to? Yes to lower the current but then I get jumps when changing travel direction from over compensation.

The ways look great and the way lube is dripping. If I run the balance program the current is consitent meaning while X is pulling 5 amps Y is pulling 7,5, so 1,5 amp over on feed moves and on idle.

I have checked motor brushes, blown out carbon dust and cleaned the commutator, but it made no difference. I just replaced the resolver on Y as I was only getting 1,65 VAC back while 1,74 from X. The new Y resolver gives back 1,75VAC, but I am still getting eggs and 3 amps on idle.

Cheers,

Andy

[alex gitzel]

make sure all the gibs are set,loose center gibs could cause this.always put the machine in e-stop and turn the screw by hand after setting gibs,it should move freely by hand,just slightly tighter than before adjustment.then reset the backlash.if you have high numbers check the end of the screw for play w/indicator they should be less than .0015 in y and .0005 in x the y is known to flex a bit more.if the numbers are high the thrust bearings should be replaced,verify the screw turns free,tight gibs will make the numbers high.the screw itself can be checked by putting the indicator tip on the screw pitch and mounted on the moving part(table,saddle)the reading should be less than .0005.you can look in the monitor cabinet door the original parameters are on a sticker on the door.you have a metric machine so the ipm will be 700.

[scadvice]

...and nice detail for future reference.

Steve

[Andy Fritz]

Hi Alex, Thanks for your advise. You may not remember but you helped me with my Z axis problems about a year ago and sent me wiring diagrams for hooking up more than one probe.

A few stupid questions;
1. I have not yet figured out how to get to the Y ballscrew to turm it by hand. There is a pipe cover surrounding the screw. Do I have to pull the motor off? I dont have tools to undo the 4 bolts as I cant get enough purchase with a standard hex wrench, (too tight space).

2. If I have never touched the Y gibs, would it not be more logical to assume there will be more play over time due to wear, rather than them getting too tight over time? Sure dirt could mess things up, but there is no visible wear on the ways and the machine spends most of it time in idle and the oil is literally pouring.

3. Is this "Angle test" not an established Fadal test program followed by tweaking pots on the amps to improve circles? I found a "Test cut" in diagnostics. It is not fully explained, but I think the idea is to diagnose problem areas. If this is the Angle test, I have already done the circle part and found that it is badly out. Now I need to know which pots to turm to fix it=).

The BL settings are done by MPG and generally my approach is just to jog and indicator up to an edge, read the indicator and position, jog away and come back to the same position. Then input BL values until the indicator reads the same, i.e. compensate for lost motion. If I set the mpg in highest resolution, move one pulse in minus travel I want to see the indicator move and move back moving the axis one pulse in plus travel. Maybe this approach gives a completely different effect running a program? I would assume that either method should be ok as long as the same method is applied to all axis?


Sorry for all the questions...

Andy

[alex gitzel]

you can get at the screw by taking the bolts off on the front y cover,pull the cover back and jog the table back,the screw is not covered there.you will need to move the cover to set the gibs anyway.there are gibs in front and back of the saddle,the x axis ones are recesed in a slot under the table on the front side,2 gibs R+L side.there are 6 gibs on the head top and bottom center and strap.adjustment is in the manual.all gibs use 1/4'' socket,exept for the x stap gibs.the gibs do wear over time and are likely to be your issue.also check your oil system the pdi pump should go to 300-400 psi on the gauge and hold for at least 5 sec.the pump will fire when you turn the machine off.if it does not hold or reach 250+psi you have a leak.i set the backlash using the mpg method and it always works for me,fadal has a program style test(in manual) if you want to see if it is the same.the amps should be tuned to the manual spec.yours sound fine 302,set them and leave them alone the issues you are having are not amp tune related, most likely machanical.i tune to spec.and do all the bl and gibs and have always got good circle cuts.the center gibs x,y,or z alow a lot of play when loose,they need to be set before looking at other things.

[Andy Fritz]

Hi Alex,

I have been busy running parts so had to wait with ripping it apart. Any way I fixed the roundness problem by ignoring the voltage and instead balancing all axis to -1 - 0 on the display. Still there is another problem with Y and that is the occasional error #18 on idle. It seems like the current draw is high and that it depends on where the saddle is positioned. For example it can be 6 amps just after cold start and moving it 0,001" to plus or minus can drop it to 2 amps. If I set the BL to a higher number than usable I can drop the current, but it is not consistent. I have removed the gibs and straps, cleaned everything, re-set to manual specs, and now the current is same or a little higher. At feed 25 IPM over the whole travel I get 6 amps in the middle, 9 amps at +10" and 10 amps at -10". I have cracked the ball nut screws open, moved as close to the motor I can, retightened and cracked open the four 3/4" bolts between casting and the bearing block and re-tightened. Is this normal current draw for the Y motor? If I remove the motor the current draw is 0,2 amp on idle. Is it important to remove the motor when aligning the ball screw? To me it looks like the current is generally high over the whole travel and the difference could be due to most wear on the ways in the middle, a little more on the Y+ as the table is usually moved to +10" at the end of part programs, but rarely moved to -10".
I have not touched the X gibs and straps and that shows 0,6 amps at CS on idle, and at F25 IPM around 2 amps in the middle and up to 3 amps at plus or minus 10" travel.

Would seem logical that the Y should pull a bit more than X due to the weight of saddle, table and fixtures, but how much more is the question.

Cheers,

Andy

[alex gitzel]

The motor must be removed when setting the y screw the weight of the motor will just pull the bearing block down and not float in as you are trying to do.move the table back then pull the motor off ( tip, use long 3/8-16 bolts with the heads cut off or threaded rod about 6-12'' long and put them in the top 2 bolt holes on the motor before taking the bottom ones out this will hold the motor up for you as you slide it out and will work as guides to reinstall.). then loosen the block and float it in,tighten and put the motor back.the amp draw will be more on the y axis because of the weight.your screw may not be straight if you left the motor on during line up and will create extra amp draw,it sounds like it may be your issue if the axis faults at idle,the screw should turn freely by hand.use one hand it will have some drag but if you can not turn it something is too tight.you may need to loosen the gibs and tighten one at a time feeling the drag as you set them,sometimes the factory spec. setting will be too tight on a worn machine.so rest the screw and then do the gibs,leave the gibs set while doing the screw if they are loose the screw will not be straight.

[Andy Fritz]

Hi Alex,

I realigned with motor off and it looks like it knocked down the current draw 0,5 amp over the whole travel. I made up a wood spacer for the Y motor as ANSI bolts are not available in Thailand. I will try to reset the gibs to specs at the ends of travel rather than in the middle. Still it would be great to have an idea of what "normal" current draw should be at idle and for example 25 ipm. Double or tripple the current draw of X?

I have also noted that the sides on the middle saddle way (where the gibs are acting) seems more dry than the top of the left and right ways. Is that normal or should one oil line lubricate the sides? If so, do I have to lift the saddle off the ways to get to some nozzle to check for clogging?

I have also re-tuned the amps and the only pot setting I am not sure about is the comp. The paper I got with the last amp-0021 said to turn it until the motor starts making a noise, then back until it just stops. Then further a full turn or something. The amp-0021 full scale is not even a full turn. When doing a circle the machine seems to slow down a lot on the quadrants. The cut looks good, but thought maybe incorrectly set comp pots could be the reason.

I also sometimes get a "jump" during feed moves of maybe 1/2" in either X or Y. Almost like a small rapid move where none should be. When this happens there is no control error and the program just continues. I have changed resolvers, cleaned out carbon dust, checked brushes and now the gibs.
Could this be bad tachs? If so how do I open and clean them without loosing magnetism? Or should I just replace them? I saw that they should output 7V per 1000 rpm. How do I command an axis motor to turn at exactly 1000 rpm to measure?

Andy

[alex gitzel]

The amp draw sounds fine in the y now.the y can draw more than x but are fairly close normaly i would have to dig out the sheet that i have to get what the normal # are.i dont test that way normaly,feeling the screw drag and backlash readings are all i normaly need.if you cant turn the screw by hand something needs adjustmet.the center y way can seem drier than the rest because the oil runs down it,if there is a oil film it should be ok,there are just holes and tubes that feed from the meter,they can be blown out with kerosene and pump if needed.the motor jump can be the tach,but i dont mess with the motors if they are bad,or in question i take a new one from my van and try it.it seems that once the motors act up they just give more trouble,so we just replace them,done,fixed,make chips and no more trouble.you should ohm the motor leads to the chassis to see if they are shorting and rebuid/replace if they are.also check the gold bullet plugs on the control cards,if loose they can cause jumping.the slow quadrant moves may be from,if you use format 1 they control reverts back to g9 regular speed machining the motors ramp down for reversal (easier on the machine and is recommended by me ). you need to command g8 high speed machining for no ramping.