Re: [Emc-users] [OT] Plasma noise resistant box
Hal Eckhart wrote: > > Thanks for responding, Jon. > > I don't really know what a commercial-grade noise filter is, > but I do have a fancy surge supressor. It never helped with > the old system much. > Hal, Sola is a company that manufactures higher end isolation transformers, surge suppressors and constant voltage transformers. I have acquired some of these from older equipment that was getting scrapped. A constant voltage transformer takes in a wide voltage range like 95-150VAC and outputs a constant 118VAC output. This is good for power line fluctuation where there are brownouts and overages from motors that brake and dump the kinetic energy back into the power lines. Just do a search on ebay. Sola's surge suppressors are going to be in the $100 range on ebay as opposed to a $20 fancy plastic one. Protecting from a one-time event ($20 unit) is different than a constant environment of transient abuse. It has been a while but I used to frequent the 3M surplus store in Minneapolis and they had all kinds of industrial surplus. I'm sure they had some Sola products. I am using a rotary phase converter on my Bridgeport VMC. I had a problem when the spindle quickly ramped down to zero during a stop spindle or a tool change. The kinetic energy in the spindle dumped back through the spindle driver and back into the power lines. Since the rotary phase converter was not as rigid as the power lines my voltage at the rotary phase converter would jump more than 15% or so. When the Bridgeport saw the voltage kick up too high it would fault and require a reset. It was a hard fault and the Z axis would fall about 0.050". A constant voltage transformer would have been a band-aid and probably would have worked in my case. Instead I decided to put a 1.5 second RC filter in the analog spindle command (0-10Vdc) line. This accomplished a couple of things for me. Spindle ramp up and down became smoother providing less wear on the spindle and drive and power fluctuations went away. > Once upon a time, I fired the plasma is the air 5 feet away > from a short piece of zip cord with a digital VOM on it. It > registered 1000 volts, which is the limit. Like you said, the energy from a plasma spark alone can radiate as electromagnetic noise and be picked up in a loop of wire some distance away. If this were my setup I would put the PC in a Faraday cage (a metal box with adequate cooling but shielded much the same way a microwave oven is). I think the Dell GX270s are mostly plastic so they are inviting EMI directly into the motherboard and I/O cards. The problem with replacing a PC whenever it dies is that the PC may issue a rapid servo command before it gives out. You can monitor EMI noise and transients by connecting a neon bulb and an LED in series. If the voltage is high enough to get the neon to ionize the neon will blip but the LED will have a brighter flash. I would connect 2 LEDs back to back and in opposite directions to detect any positive and negative transients. You can take an old PC and connect the neon bulb & LED setup from any input or output you want to monitor and ground of the PC. If enough voltage is induced in an encoder line, etc. the neon bulb and one of the LEDs should blip. The transients may also be too fast or too low in energy to see anything but just enough to cause damage. If transients are getting into the PC through I/O lines, the Faraday cage won't help much. In a super noisy environment I would personally run fiber optic transceivers between the PC in a Faraday cage and the machine. Now that optical S/PDIF cables are a couple of bucks on ebay you can run 15' of plastic fiber between your PC and the rest of the world. The trick is to find a S/PDIF transceiver board with enough I/O to completely pass all the I/O back and forth to the PC. You might have to make a pair of custom isolation PC boards. Dennis J. Deyen Product Design Mgr. Pedersen Power Products 3900 Dahlman Ave. Omaha NE 68107 - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] CNC design questions
Rafael, This is an idea on how to provide linear motion to a single axis: http://www.oemdynamics.com/hld_animation/hld_intro.html The Harmonic Linear Drive (HLD) article was posted in Machine Design on 6/16/08: http://machinedesign.com/ContentItem/719/72741/ScanningforIdeasNovelRecirculatingBeltPowersLinearActuator.aspx The large pulley and small pulley sets are connected together. This provides a speed difference at the outer radius of the pulleys. If the large and small pulleys were the same size then their circumference speed difference would be zero and the carriage would not move. If the smaller pulley had 49 teeth and the large pulley had 50 teeth the pulley edge speeds would be 49 and 50 respectively. This would make the carriage move at a rate of 1-(49/50) = 0.02 or 2% of the rate of the large pulley edge radius speed. I have always been a fan of rotary harmonic drives due to their "near zero backlash" and the example above is just a linear version of it. The linear system provides fast acceleration using low mass components although belt speeds can be high compared to the carriage speed (30-50x faster). If the timing belt in a Honda type R engine can accelerate from 700 RPM to 9000 RPM in 1/2 sec then a properly selected belt should work fine in a linear motion control application. I think the HLD system could work well with a direct drive DC motor because a DC motor can accelerate quickly to 3000 RPM and provide 560" per minute of travel speed with 3" dia pulleys (49 tooth & 50 tooth). My Bridgeport VMC has max travel speeds of 300 IPM. It is far from being state-of-the-art but gives me a baseline for other projects. High quality DC motors with built in encoders can be cheap if you look in the right place. I have quite a few Ametek drive motors from old tape drives that I picked up for $5-10 ea. at a local army surplus store. I'm a sucker for high quality motion control that's nearly free. Dennis - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] diesel motor control
It is amazing how many similar interests there are in the industrial community. I have a functional Bridgeport VMC in the garage and a Makino wire EDM that needs a little maintenance before it is fired up. I'm not really interested in something unless I think I can build something myself. In the driveway is an '85 Chevy 6.2L diesel flatbed. I have about $1200 in it right now. The diesel technology from '85 is much different than today. It runs fine but does not have forced air induction and belches black smoke. Back then they didn't care much about air-fuel ratios with diesels. I did quite a bit of research on diesels mostly in the racing arena (torque = speed). Bosch does have 32,000PSI injectors but they can be fed with a relatively low pressure pump of 1500PSI. The fuel is compressed in the injector using a piezo stack. Control of the piezo stack is the key. There is probably a science behind controlling the new direct injection (DI) systems. Some mfgs are known to pressurize the cylinder and inject a minimal drop of fuel in a small auxiliary pocket of the combustion chamber. As the fuel ignites and starts to spread the flame front into the main chamber a second and longer pulse of fuel is injected and probably at top dead center or after (power pulse). The result is a much quieter explosion by controlling the volume of fuel over time. A standard volume of fuel in a power stroke thrown in all at once yields a bang (old style knock). A small puff of an explosion accompanied by a longer pulse width softens the blow but the expansion of gases still provides a lot of torque. The small volume of fuel from DI is also compatible with propane injection because the propane is looking for any ignition source, large or small. This would also help fuel economy. I wouldn't be opposed to putting a 10 gal propane tank in the trunk. I like diesel because it is simple. Put a drop of oil in an environment with a high compression ratio at 150 degrees F and boom. The newer Bosch direct injection (DI) injectors can output as little as 1 mm^3 of diesel by sending a narrow pulse width to the injector. You can virtually figure your fuel economy based on some assumptions: 1 mm^3 fuel every other revolution per cylinder in a 4-cycle engine, 1 mm^3 of fuel per cylinder is enough to provide sufficient torque to overcome wind and rolling resistance (40HP typ), diesel produces 2x the torque per firing than gas and a light weight chassis typically running at 60MPH highway speed. Instead of using a rotary encoder on the crank I would use a 4-6-8 position sensor that had fairly accurate geometry cylinder to cylinder and key off that. The engine is not going to change speed so quickly that you can't predict where it is. From one cylinder firing to the next firing the engine won't double in speed. Diesels accelerate slowly. I would have predictive software that knew the current speed and acceleration of the crank. A combination of this data and the position sensor would tell me when to fire. As long as I am consistent in timing cylinder to cylinder the engine won't care. The DI injectors allow you to tweak before, during and after top dead center fuel pulses. Diesel is a slow burning fuel. At higher speeds you would need to advance the DI pulse to achieve a near complete burn by the end of the cycle. At 5000 RPM the combustion period is about 6ms. I'll probably never get to build my 2L 4-cyl 5000RPM 300HP diesel racing engine but some day may buy a DI injector for $300 or so and retrofit my 21" JD lawnmower with a diesel. Dennis - Original Message - From: "Stuart Stevenson" <[EMAIL PROTECTED]> To: "EMC2-Users-List" Sent: Friday, November 30, 2007 11:51 AM Subject: Re: [Emc-users] diesel motor control > Gentlemen, >I am like the 5 year old boy with a hammer. 'Everything' looks > like a nail. I want to control the injectors with EMC. >The current diesel technology is common rail injection using HIGH > injection pressure. Bosch talks about 2200 bar. 2200 bar is almost > 32,000 psi. WOW! >They also have multiple injections per combustion cycle. >They also talk about variable geometry during the injection cycle. >This sounds like a job (nail) for EMC. >I won't try for the high pressures as I will use currently > available injectors and pumps. >To get the project going I need to start simple. >I will use a used pump and injectors from a Volkswagon (or other). > I will modify the head to accept the injectors. I will build an > adaptor or bracket to mount the pump. >Voila! :) running in two weeks! heh heh heh > thanks > Stuart > > - > SF.Net email is sponsored by: The Future of Linux Business White Paper > from Novell. From the desktop to the data center, Linux is going > mainstream. Let it simplify your IT future. > http://altfarm.mediaplex.com
Re: [Emc-users] Paging doctor P.I.D. Guru
Hello, I have solved a similar problem using 10MHz+ AB encoder counts feeding into a microcontroller. The microcontroller is a 20 MIPS little thing that has limitations on the interrupt rate and/or its counter-timer section. I think 4MHz was the max input frequency. My solution used a CPLD to track the actual position and give the microcontroller a prescaled count to use at higher speeds and also provide the lower un-prescaled count bits for position feedback when working at lower speeds. The prescaler divided the encoder position by 32 so the maximum count the microcontroller would see was 10MHz/32 = 312kHz. The counter section I/O was capable of reading this. When not at top velocity the microcontroller would sense the scaled speed was below a certain threshold and begin to use the lower 5-bits of the un-prescaled CPLD count for position feedback. You may do something similar by prescaling your encoder count and using that for EMC2. I'm guessing with that high of a resolution the lower bits can simply be thrown away. Dividing by 16 in your case would get below the threshold required for your EMC2 setup. Dennis - Original Message - From: "Marc van Doornik" <[EMAIL PROTECTED]> To: Sent: Friday, November 23, 2007 2:06 AM Subject: [Emc-users] Paging doctor P.I.D. Guru > Any PID tuning gurus are welcomed to shed their thoughts on a bit of a > controlling issue I'm having. > > Recently, we have upgraded a functioning EMC2-controlled laser cutter > with a new motion stage. Initially, we used stepper motors, connected > via Geckodrive G201 drivers. Now, the X/Y-stage is a state-of-the-art > linear motor-driven job. Some relevant info: > > * Linear motors, peak force around 700 N > * Max. velocity 2 m/s or about 79 inches per second > * Max. acceleration 3 g > * Matching PWM amplifier with direct commutation input and three-phase > output > * Optical encoders with 100 nm (~0.04 inch) resolution and > differential output > * Precision linear guides (no air bearing due to contamination issues), > so there is a bit of friction involved > * 300 mm (~12 inch) stroke length > * Motenc-lite servo card > > You can probably imagine my nervousness and reluctance to try and tune > such a monster by feeding it a stepped input for getting a nice PID > response. You really don't want this thing to go instable on you, trust > me. From one end of the stage to the other takes it around 150 ms. > Seriously scary stuff, especially with the forces involved. > > So far, connecting everything up has been easy as pie. Sort of, anyway. > I've written a HAL component to generate the proper phases for direct > commutation from a velocity input. When controlling the motors in an > open loop, the velocity does match the input fairly well and the > encoders give the correct position. In that respect, all is well. > > Now for my actual question: what would be the best control and tuning > strategy? I suppose I'll have to cascade a velocity loop and a position > loop to get sufficient stiffness (given the spring-like behaviour of a > linear motor), but I'm not sure how to do this, especially as far as > tuning is concerned. My idea is as follows: > > 1. Create an expanded version of the three-phase component with an > acceleration input (internally integrated to get the velocity signal). > Feed back the encoder data to generate a position signal and a velocity > signal. This way, the output of the velocity PID controls the > acceleration (like it should, am I correct?), the velocity signal from > the encoder being fed back into the velocity PID. > > 2. Tune the velocity loop - but how? > > 3. Add another PID before the velocity input to control the position and > feed it the position data of the encoder. > > 4. Tune position loop. > > One problem that springs to mind is the spring-like behaviour of the > motor. This manifests itself as a hysteresis in the output position due > to the friction of the roller guides (about 15 N). Obviously, this > hysteresis should be compensated for as fast as possible, preferably > with the maximum velocity of the stage (i.e. 2 m/s). However, the > maximum velocity the servo card can keep up with (because of encoder > resolution) is 200 mm/s. Should I try to circumvent this by applying > feed forward of some sorts? Maybe I'm better off adding a rough backlash > compensation in the three phase component itself? > > Any insights on this matter are greatly appreciated. > > TIA, > > Marc. > > > - > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ > ___ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > - Th
