Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
- MachCNC
- Messages
Search
Re: RC servo signal - R/C brushless / brushed spindle motor speed controller
Hugh Prescott
Do the RC motor controllers use 1mS as off and 2ms full throttle, orThe transmitters can be set either way to match up with the speed controller. I have never seen a standard published so you should probabbly design for either. Some do not have reverse and use the full span for forward, most are adjustable by using the ESC programing function built into most of the new ones. As I understand, a connection to the servo , or motor controller isby a 3 wire connector, Gnd, +5V and the control signal. For the motorthe +5V supplies the servo power. For a motor controller, my guess is that itEvery speed controller I have tested or serviced has SUPPLIED 5 - 6 volts on the +5 wire to run the receiver and other servos. This feature is called a BEC (Battery Elimnator Circut). Amp capacity ranges up to 1 amp on some of the new ESCs. You do not need to (you may damage the ESC or BEC circut if you do) provide 5 volts to the ESC on the control cable just ground and the standard control signal. On our bench testers (my design, not commercially available) at work (we own a large hobby shop that does lots of repair work) I installed a separate lead for testing ESC's that does not provide 5 volts but instead applies a variable load and displays the BEC output voltage. Servo output pulse is a standard 5 volt pulse of 1 - 2 MS occuring every 16 MS. Hugh |
IMS IM483 Vs. Compumotor OEM 650 controlers
Hi everyone,
I was hoping to pick someones brain on this. I have two 3 axis motor controllers, one is made up of IMS IM483's and the other OEM 650's. The IMS is using a 40vdc power supply and the compumotor is using thrie 75 vdc OEM 300 power supply all are set to the same resolution. Now for the question both units turning the same motor the IMS runs faster and smoother than the OEM 650 the motor's are Vexta 1.4 amp nema 34's. From everything I have read the OEM 650 at 75vdc should run the motor faster than the IMS IM483. With the IMS I get about 50 IPM and with the OEM 650 I get less than 12 IPM and lots of chatter noise. Any clues? Using Mach 3 latest lockdown release and the same computer to run both, just unplug one and plug up the other. Thanks Mike Shell |
Re: RC servo signal - R/C brushless / brushed spindle motor speed controller
Peter Homann wrote:
Hi Manu,I think some of these will be a problem unless you can write a macro to cycle the RC pwm to reset the speed controller out of failsafe. They tend to have a safety failsafe which requires the sticks to be moved to various positions until bleeps are heard, only then is the device armed. Some of these motors and controllers can produce wicked power! A friend of mine competes at world level in f5b, and those brushless aveox motors etc with 30 cells produce several horsepower. I may speak to a friend who designs his own brushed and brushless controllers. He may be able to come up with something more suitable to the cnc cause. Wayne.... -- --------------------------------------------------------------------------- Wayne M Weedon Email: wayne@... Fdos Design Poole UK Tel +44-1202-677025 Fax +44-1202-770515 Mobile: 07774 439915 Specialists in small batch & Production Mechanical/Electrical Engineering --------------------------------------------------------------------------- |
Re: RC servo signal - R/C brushless / brushed spindle motor speed controller
Wayne,
toggle quoted message
Show quoted text
I'm sure you are correct. From memory, it is also how they set the off position as well. Cheers, Peter. Wayne Weedon wrote: Peter Homann wrote:Hi Manu,I think some of these will be a problem unless you can write a macro to cycle the RC pwm to reset the speed controller out of failsafe. --
------------------------------------------------------------------ Web: www.homanndesigns.com email: homann@... Phone: +61 421 601 665 www.homanndesigns.com/ModIO.html - Modbus Interface Unit www.homanndesigns.com/DigiSpeedDeal.html - DC Spindle control www.homanndesigns.com/TurboTaig.html - Taig Mill Upgrade board |
Re: Driver problem new computer
elaurijsen
Hi,
I just reinstalled Windows XP in the normal way, so with ACPI. Now Mach 3 is running ;-) It is not as smooth yet as i want but i didn't use the optimisation-file yet. Happy that is working again! Thanks for your help. Cheers, Eric --- In mach1mach2cnc@..., art <fenerty@...> wrote: ACPI) i think?The LPT-port is running on 378 (=standard value)
|
Re: RC servo signal - R/C brushless / brushed spindle motor speed controller
Hi Manu,
toggle quoted message
Show quoted text
I've had a look and I'm fairly sure that I can get the DigiSpeed-XL produce an RC signal that should work fine. I can add it as a mode selected by the mode jumpers. I have a question. Normally a pulse of 1.5ms centres the RC servo. 2mS is full CW. 1mS is full CCW. Do the RC motor controllers use 1mS as off and 2ms full throttle, or 1.5mS is off, 2mS is full throttle, and 1mS is full reverse. I ask this as there is the direction line into the DigiSpeed-XL. I can use this to put the motor controller into reverse (< 1.5ms) Or as I suspect, do some motor controllers use 1mS as off with no reverse and others operate as described as above? As I understand, a connection to the servo , or motor controller is by a 3 wire connector, Gnd, +5V and the control signal. For the motor controllers, how much current does the +5V connector draw? For a standard servo, the +5V supplies the servo power. For a motor controller, my guess is that it supplies only a reference or just small amount for the logic circuitry. From my research, I can add a 3 pin header connector that the 3 pin servo cable can be plugged into. BTW, what sort of motor do the RC motor controller drive. I thought they were just small 400 series motors? I can probably have one ready in a couple of weeks if you are interested. Cheers, Peter. Manu CNC wrote: Hi Art, --
------------------------------------------------------------------ Web: www.homanndesigns.com email: homann@... Phone: +61 421 601 665 www.homanndesigns.com/ModIO.html - Modbus Interface Unit www.homanndesigns.com/DigiSpeedDeal.html - DC Spindle control www.homanndesigns.com/TurboTaig.html - Taig Mill Upgrade board |
Re: Slow computer
Steve Blackmore
On Sat, 03 Jun 2006 07:16:37 +0100, you wrote:
Spent some time yesterday playing with a TaigCNC machine and comparing May not be the processor - could be the chipset, especially if it uses the same Via chipset as in some models of Shuttle. Some work, some just don't :( Steve Blackmore -- |
Re: Slow computer
Lester Caine wrote:
Right guys ( and galls? )Lester. What I have heard, but not tried it myself is that the Mini-ITX motherboards with Intel Processors are ok. The Via ones have problems. Maybe someone else will verify. I'd expect they would all be ok with a G100. Wayne... -- --------------------------------------------------------------------------- Wayne M Weedon Email: wayne@... Fdos Design Poole UK Tel +44-1202-677025 Fax +44-1202-770515 Mobile: 07774 439915 Specialists in small batch & Production Mechanical/Electrical Engineering --------------------------------------------------------------------------- |
Re: Slow computer
Lester Caine
John PRENTICE wrote:
--- In mach1mach2cnc@..., Lester Caine <lester@...> wrote:Currently no network stuff enabled! When I get the machine over here I'll run things up on the scope ;)What *IS* the slowest processor speed that will work cleanly - or is the less than smooth pulse generation a 'facet' of using windows?No you can get immaculate results 1000s of users demonstrate that. Less than regular pulse train will affect finish, but if people are saying that should not happen ...Given the quality of work that the machine CAN produce, any pattern to the axis movement is likely to affect the surface finish.Processor speed not likely to affect finish etc. If you run out of cycle then it all chokes up. GUI goes slow (or stopped). 1 GHZ is certainly OK for me but struggles running Modbus and a big Macropump.I'd probably be looking at a bigger machine if I wanted more inputs, but that is the point when G100 comes into the equation anyway. -- Lester Caine - G8HFL ----------------------------- L.S.Caine Electronic Services - Model Engineers Digital Workshop - Treasurer - Firebird Foundation Inc. - |
Re: Slow computer
Lester Caine
davidimurray wrote:
Have you checked the parallel port output voltage? A lot of laptops and modern desktops now seem to put out only 3.8V instead of the 5V your driver board probably needs.I'm seeing 5V, but as I said to John, I'll check that on the scope when I have the machine in the workshop here. The thing I'm trying to sort out is what problems will happen if I'm just supplying the MicroMill without a computer, and the DOS package is less demanding than Mach ;) -- Lester Caine - G8HFL ----------------------------- L.S.Caine Electronic Services - Model Engineers Digital Workshop - Treasurer - Firebird Foundation Inc. - |
Re: Slow computer
Replies intersperesed below
--- In mach1mach2cnc@..., Lester Caine <lester@...> wrote: <snip> Even slowing things down, running just 25kHz and slower feed rates we have adecided 'pattern' to the noise from the stepper, like something is sticking every coupleof seconds, while the DOS has none even at a faster rate.I would suspect a periodis Tick in motion is Etherner card set to autonegoitiate 10/100. Try setting it to 10. What *IS* the slowest processor speed that will work cleanly - or is the lessthan smooth pulse generation a 'facet' of using windows?No you can get immaculate results 1000s of users demonstrate that. Given the quality ofis likely to affect the surface finish.Processor speed not likely to affect finish etc. If you run out of cycle then it all chokes up. GUI goes slow (or stopped). 1 GHZ is certainly OK for me but struggles running Modbus and a big Macropump. John Prentice |
Re: Slow computer
Hi Lester
Have you checked the parallel port output voltage? A lot of laptops and modern desktops now seem to put out only 3.8V instead of the 5V your driver board probably needs. Cheers Dave --- In mach1mach2cnc@..., Lester Caine <lester@...> wrote: comparing Mach3 with the DOS package that normally drives it. The originalcontrol program is running on an old machine that would probably not evenload W2k so I had taken along an ITX box with W2k already installed.Only problem is that it is an 800MHz C3 processor, but since this isintended to be an embedded computer to go with the machine nothing else wasthe same tests in Mach3 give a number of problems. Even slowing thingsdown, running just 25kHz and slower feed rates we have adecided 'pattern' to the noise from the stepper, like something is sticking everycouple of seconds, while the DOS has none even at a faster rate.fast enough? Neither of my laptops have parallel ports, so I could nothook them up, and the desk machines are rack mounted, so until I canget a machine over here it's difficult to try a faster processor. What*IS* the slowest processor speed that will work cleanly - or is theless than smooth pulse generation a 'facet' of using windows? Given thequality of work that the machine CAN produce, any pattern to the axismovement is likely to affect the surface finish.processor with each machine but I am wondering if that will work, or if I amstuck with just running DOS and leaving Mach until the G100 or somethingsimilar is a cost effective alternative :(faster processor ;) |
Re: RC servo signal - R/C brushless / brushed spindle motor speed controller
Manu CNC
RC-Servo signal:
toggle quoted message
Show quoted text
Manu (@manu, manutoys, vantiv) -----Original Message-----
From: mach1mach2cnc@... [mailto:mach1mach2cnc@...]On Behalf Of Manu CNC Sent: Saturday, June 03, 2006 9:39 AM To: mach1mach2cnc@... Subject: RE: [mach1mach2cnc] RC servo signal - R/C brushless / brushed spindle motor speed controller Hi Art, My setup : Kernel: 25,000Hz Spindle Base Freq: 50Hz Min, - Max Pully ration: 1 to 10,000 After scope measurements I get this ! If I set the scope resolution to 2ms/div I do get the correct signal!! _|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|__ _|_ 2ms/div 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 1 2 3 rpm 200 _||______________________________________||__________________________ 300 _||____________________________________||________________________ 400 _||___________________________________||________________________ BUT, if I zoom out to 0.2 SECONDS/div I get this: distance between the pulses 1.5 SECONDS !!! Zooming into a pulse, I get the signal as above! rpm 200 _||______________________________________||_________________________________ _________ 4000_||___________________||__________ ______________ 9000_||| |_____ Mybe my scope is confusing me Anyway, I also attached a RC-servo motor to it and it moves in an accelerated way to one extreme, whatever rpm speed setting: start slowly and accelerate at the end. I can confirm that an RC-PWM-signal are Pulses of 1 to 2 msec long repeated every 20-30 msec! Manu (@manu, manutoys, vantiv) -----Original Message----- From: mach1mach2cnc@... [mailto:mach1mach2cnc@...]On Behalf Of art Sent: Saturday, June 03, 2006 1:03 AM To: mach1mach2cnc@... Subject: Re: [mach1mach2cnc] RC servo signal - R/C brushless / brushed spindle motor speed controller Manu: Well, if you select 50hz as the base frequency, and set the Max Pulley ratio to 10,000 and only select speeds from 200 - 400RPM thats abotu what you woudl get. A 1 - 2ms pulse every 20ms. :) Thanks, Art www.artofcnc.ca Videos And Support Forums Users Map: ----- Original Message ----- From: "Manu CNC" <mansys@...> To: <mach1mach2cnc@...> Sent: Friday, June 02, 2006 6:04 PM Subject: RE: [mach1mach2cnc] RC servo signal - R/C brushless / brushed spindle motor speed controller Grrrrrrr,_|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|__ _|_ 2ms/div_||___________________||______________ __________100%_||| |_____ www.machsupport.com - Web site Access Yahoo! Groups Links www.machsupport.com - Web site Access Yahoo! Groups Links |
Re: RC servo signal - R/C brushless / brushed spindle motor speed controller
Manu CNC
Hi Art,
toggle quoted message
Show quoted text
My setup : Kernel: 25,000Hz Spindle Base Freq: 50Hz Min, - Max Pully ration: 1 to 10,000 After scope measurements I get this ! If I set the scope resolution to 2ms/div I do get the correct signal!! _|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|__ _|_ 2ms/div 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 1 2 3 rpm 200 _||______________________________________||__________________________ 300 _||____________________________________||________________________ 400 _||___________________________________||________________________ BUT, if I zoom out to 0.2 SECONDS/div I get this: distance between the pulses 1.5 SECONDS !!! Zooming into a pulse, I get the signal as above! rpm 200 _||______________________________________||_________________________________ _________ 4000_||___________________||__________ ______________ 9000_||| |_____ Mybe my scope is confusing me Anyway, I also attached a RC-servo motor to it and it moves in an accelerated way to one extreme, whatever rpm speed setting: start slowly and accelerate at the end. I can confirm that an RC-PWM-signal are Pulses of 1 to 2 msec long repeated every 20-30 msec! Manu (@manu, manutoys, vantiv) -----Original Message-----
From: mach1mach2cnc@... [mailto:mach1mach2cnc@...]On Behalf Of art Sent: Saturday, June 03, 2006 1:03 AM To: mach1mach2cnc@... Subject: Re: [mach1mach2cnc] RC servo signal - R/C brushless / brushed spindle motor speed controller Manu: Well, if you select 50hz as the base frequency, and set the Max Pulley ratio to 10,000 and only select speeds from 200 - 400RPM thats abotu what you woudl get. A 1 - 2ms pulse every 20ms. :) Thanks, Art www.artofcnc.ca Videos And Support Forums Users Map: ----- Original Message ----- From: "Manu CNC" <mansys@...> To: <mach1mach2cnc@...> Sent: Friday, June 02, 2006 6:04 PM Subject: RE: [mach1mach2cnc] RC servo signal - R/C brushless / brushed spindle motor speed controller Grrrrrrr,_|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|__ _|_ 2ms/div_||___________________||______________ __________100%_||| |_____ www.machsupport.com - Web site Access Yahoo! Groups Links |
Slow computer
Lester Caine
Right guys ( and galls? )
Spent some time yesterday playing with a TaigCNC machine and comparing Mach3 with the DOS package that normally drives it. The original control program is running on an old machine that would probably not even load W2k so I had taken along an ITX box with W2k already installed. Only problem is that it is an 800MHz C3 processor, but since this is intended to be an embedded computer to go with the machine nothing else was installed on it. Operation from the DOS program is smooth and clean, but running the same tests in Mach3 give a number of problems. Even slowing things down, running just 25kHz and slower feed rates we have a decided 'pattern' to the noise from the stepper, like something is sticking every couple of seconds, while the DOS has none even at a faster rate. Is this likely to be due to the fact that the processor is not fast enough? Neither of my laptops have parallel ports, so I could not hook them up, and the desk machines are rack mounted, so until I can get a machine over here it's difficult to try a faster processor. What *IS* the slowest processor speed that will work cleanly - or is the less than smooth pulse generation a 'facet' of using windows? Given the quality of work that the machine CAN produce, any pattern to the axis movement is likely to affect the surface finish. Ideally I would like to include a 1.2 or 1.5Ghz ITX based processor with each machine but I am wondering if that will work, or if I am stuck with just running DOS and leaving Mach until the G100 or something similar is a cost effective alternative :( I will leave the other problems until I can run things with a faster processor ;) -- Lester Caine - G8HFL ----------------------------- L.S.Caine Electronic Services - Model Engineers Digital Workshop - Treasurer - Firebird Foundation Inc. - |
Re: Driver problem new computer
Steve Blackmore
On Fri, 02 Jun 2006 23:25:29 -0000, you wrote:
Just forgot to say that i installed the newest version of Mach3 andYes, that and a HT model. Also its not good idea to disable ACPI or anything unless you really HAVE to. The "optimisation" file should have a disclaimer saying "only do this as a last resort". Steve Blackmore -- |
Re: amount of steps for servo
--- In mach1mach2cnc@..., "mlaws1172" <mlaws1172@...> wrote:
Your (current) number should be based on the highest target speed (in IPM) calculated to how many pulses per second it takes to do that. It's a function of the "gearing" and the line count of the encoders. Encoders with higher line counts require more pulses per second to go a certain speed. Run the numbers and plug in the max 45,000 pulses per second that the parallel port version does. Wildcard: Gecko's and Rutex have built-in step multipliers that can be used to get around the limits. Basically the raw resolution (not particulary the accuracy) of a machine is the smallest distance one pulse will move it. Obviously the smaller the distance the better the resolution HOWEVER unused resolution is just wasted money. A plasma tabel with .005 accuracy (considering all the mechanical inaccuracies) is sufficient. A resoltuion of greater than 10 times the final accuracy is plenty. Most builders find that motors with 250 to 500 lines (1000 to 2000 pulses per rev with any level of belt reduction gives them plenty of resolution. Encoders with higher counts quickly make the math such that either faster pulse rates than parallel can generate OR step multipliers have to be introduced. Example. A servo with a 1000 line encoder puts out 4000 pulses (in quadrature) each rev of the shaft. If we use rack and pinion and a gear that moves 1" per rev (easy math). If the belt reduction is a common 4:1 the pulse count per inch is 16,000. Lets say our target is 300 Ipm (5 IPS) then we have to put out 80,000 pulses per second to get there....well past what we can provide with a parallel port. With the above system your resolution is .00000625 inches. Lets say the motor is rated at 3000 RPM so a belt reduction of 4:1 makes that 750 PRM and since it moves one inch that equates to 750 IPM. I won't go further but you have several choices and each one effects other parameters. |
Re: R/C brushless / brushed spindle motor speed controller
nattyone960
I think the Phoenix type controllers are already opto isolated, but I
am not sure. According to the advice given by a member on CNC-Toolkit rigging a Servo Tester to provide a pulse should allow speed control of the motor. --- In mach1mach2cnc@..., Peter Homann <groups@...> wrote: the RC signal of the boards test points, before the Isolation circuitry.signal or does the RC speed controller already have opto isolation?setting 900the PWMbase to 50. will(45,000/50) steps in your PWM signal. athave45 steps (900 steps/20ms) per ms. haveas 45pulley settings on your machine. Probably better than what you getnow,but not a great improvement. max.Machto generate 1ms PWM pulse for zero speed, and a 2 ms pulse for to produce--the signal if there is enough interest. |
Re: amount of steps for servo
art
Hi Mike:
toggle quoted message
Show quoted text
Unimportant. It will work with any amount. Thanks, Art www.artofcnc.ca Videos And Support Forums Users Map: ----- Original Message -----
From: "mlaws1172" <mlaws1172@...> To: <mach1mach2cnc@...> Sent: Saturday, June 03, 2006 12:29 AM Subject: [mach1mach2cnc] amount of steps for servo Seeing how i haven't asked any stupid questions for a while, i thought it was time |
to navigate to use esc to dismiss