Re: [Emc-users] stepgen pwmgen fixes
Sebastian Kuzminsky wrote: Eric (and anyone else who's had problems with pwmgen), please try this: http://highlab.com/~seb/bzr/emc2/hostmot2/src/hal/drivers/mesa-hostmot2/pwmgen.c I just committed this fix to the CVS TRUNK (along with pwmgen.frequency and some other minor stuff). Let me know if this fixes pwmgens for you guys. -- Sebastian Kuzminsky Theo: Julian? I haven't seen you in twenty years. You look good. The picture the police have of you doesn't do you justice. Julian: What do the police know about justice? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Sebastian, Can you check to see that the pwmgen.enable is working properly. I was running samples all day (which looked great, btw, after resolving a very odd problem) so I did not have a chance to do much testing. I am now dropping the enable when the shutter closes, however the pwmgen.value is still getting values. The laser was still coming on even when the enable was low. I will try setting the frequency of the pwm from the latest update and hopefully test that tomorrow. Regards, Eric I bet that's it then. The hm2_5i20 pwmgen currently goes at 25 KHz with no way to change it. I'll add a HAL parameter and let's see if that helps. - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
On Tue, 23 Sep 2008, Eric H. Johnson wrote: Date: Tue, 23 Sep 2008 21:09:44 -0400 From: Eric H. Johnson [EMAIL PROTECTED] Reply-To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net To: 'Enhanced Machine Controller (EMC)' emc-users@lists.sourceforge.net Subject: Re: [Emc-users] stepgen pwmgen fixes Sebastian, Can you check to see that the pwmgen.enable is working properly. I was running samples all day (which looked great, btw, after resolving a very odd problem) so I did not have a chance to do much testing. I am now dropping the enable when the shutter closes, however the pwmgen.value is still getting values. The laser was still coming on even when the enable was low. I will try setting the frequency of the pwm from the latest update and hopefully test that tomorrow. Regards, Eric Well I'm not Sebastian, but I know why you are seeing that behavior: The enable register was designed for our PWM and Analog output daughter cards so it doesnt actually affect the PWM output, its just a group of (active low) output bits used to enable the daughter card output channels. I can certainly change this behavior so that the PWM output is forced low when a channel is not enabled (cheap in terms of FPGA hardware). The active low enable output is used to guarantee that the motor drive outputs are off at power up (or watchdog bite) I dont think there are any bad side effects of forcing the PWM output low if disabled, but there are possible bad side effect of using a active high PWM output without the enable (Laser is on high at power up unless you have some other active low enable signal) Peter Wallace Mesa Electronics (\__/) (='.'=) This is Bunny. Copy and paste bunny into your ()_() signature to help him gain world domination. - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Peter C. Wallace wrote: On Tue, 23 Sep 2008, Eric H. Johnson wrote: Sebastian, Can you check to see that the pwmgen.enable is working properly. I was running samples all day (which looked great, btw, after resolving a very odd problem) so I did not have a chance to do much testing. I am now dropping the enable when the shutter closes, however the pwmgen.value is still getting values. The laser was still coming on even when the enable was low. I will try setting the frequency of the pwm from the latest update and hopefully test that tomorrow. Well I'm not Sebastian, but I know why you are seeing that behavior: The enable register was designed for our PWM and Analog output daughter cards so it doesnt actually affect the PWM output, its just a group of (active low) output bits used to enable the daughter card output channels. I can certainly change this behavior so that the PWM output is forced low when a channel is not enabled (cheap in terms of FPGA hardware). The active low enable output is used to guarantee that the motor drive outputs are off at power up (or watchdog bite) I dont think there are any bad side effects of forcing the PWM output low if disabled, but there are possible bad side effect of using a active high PWM output without the enable (Laser is on high at power up unless you have some other active low enable signal) The driver firmware currently communicate the pwmgen.XX.enable pin to the /Enable IO pin on the IO connector (I just re-checked this). Peter's got a good argument for why the value of the PWM pin shouldnt matter to the downstream equipment if the /Enable IO pin is high. Eric, how are you handling the start-up condition (and potentially the watchdog condition) that Peter is describing? What would it mean to keep the PWM IO pin low if the .enable HAL pin is low? Should the FPGA *not* let the PWM pin go to high-impedance/pullup at bootup and at watchdog bite? -- Sebastian Kuzminsky Hey look, it's a man! A meatie! Look at the meatie! Hi meatie! How's the weather? -- from Cathedrals by Alex Hardison http://www.flurb.net/5/5hardison.htm - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
On Tue, 23 Sep 2008, Sebastian Kuzminsky wrote: Date: Tue, 23 Sep 2008 20:24:42 -0600 From: Sebastian Kuzminsky [EMAIL PROTECTED] Reply-To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net Subject: Re: [Emc-users] stepgen pwmgen fixes Peter C. Wallace wrote: On Tue, 23 Sep 2008, Eric H. Johnson wrote: Sebastian, Can you check to see that the pwmgen.enable is working properly. I was running samples all day (which looked great, btw, after resolving a very odd problem) so I did not have a chance to do much testing. I am now dropping the enable when the shutter closes, however the pwmgen.value is still getting values. The laser was still coming on even when the enable was low. I will try setting the frequency of the pwm from the latest update and hopefully test that tomorrow. Well I'm not Sebastian, but I know why you are seeing that behavior: The enable register was designed for our PWM and Analog output daughter cards so it doesnt actually affect the PWM output, its just a group of (active low) output bits used to enable the daughter card output channels. I can certainly change this behavior so that the PWM output is forced low when a channel is not enabled (cheap in terms of FPGA hardware). The active low enable output is used to guarantee that the motor drive outputs are off at power up (or watchdog bite) I dont think there are any bad side effects of forcing the PWM output low if disabled, but there are possible bad side effect of using a active high PWM output without the enable (Laser is on high at power up unless you have some other active low enable signal) The driver firmware currently communicate the pwmgen.XX.enable pin to the /Enable IO pin on the IO connector (I just re-checked this). Peter's got a good argument for why the value of the PWM pin shouldnt matter to the downstream equipment if the /Enable IO pin is high. Eric, how are you handling the start-up condition (and potentially the watchdog condition) that Peter is describing? What would it mean to keep the PWM IO pin low if the .enable HAL pin is low? Should the FPGA *not* let the PWM pin go to high-impedance/pullup at bootup and at watchdog bite? The PWM pin would only be forced low when the channel is disabled in the PWM enable register. After Watchdog bite and at bootup, all I/Os would be in the high state. Still think best is to use the active low PWM enable output as separate hardware enable, perhaps moved to one of the GPIO pins if a 7I37 is used to enable the laser. Peter Wallace Mesa Electronics (\__/) (='.'=) This is Bunny. Copy and paste bunny into your ()_() signature to help him gain world domination. - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Sebastian, I am not sure what you mean by the FPGA register. Do you mean the interface register between the driver and the m5i20, or the register actually handling the PWM within the FPGA? As I said in an earlier email (off list), I concur that the value being sent to the FPGA through the interface register is correct (providing the bit shift to bits 16-27 is where the value is supposed to go), which would confirm that my pwmgen.scale is set to 1.0. I was able to measure the duty cycle on a pain old multi-meter. At values of 0.125 and below, the voltage would start dropping from the almost 100% duty cycle voltage of just under 3.3V, and very linearly for the range of 0.125 down to 0.0. The voltage reading from my meter was in fact very accurate. Here are the voltage readings I got: ValueVoltage =0.126 3.278 0.125 3.272 0.120 3.144 0.110 2.884 0.100 2.623 0.090 2.362 0.080 2.102 0.070 1.841 0.060 1.580 0.050 1.320 0.040 1.059 0.030 0.798 0.020 0.538 0.010 0.278 Regards, Eric How are you measuring duty cycle? I'm confident that (at least in my test setup) setting pwmgen.XX.value causes the correct number to be set in the FPGA register. Take a look at http://pastebin.ca/1208030. I updated pwmgen.c to decode the number in the PWM Value Register, then used raw.dump-state to show the register when pwmgen.value was -1.0 to +0.3 in increments of 0.1. Lines 1-22 show the relevant dmesg output for .value=0, then I just show the PWM Value Register for the other .value settings. In addition to the correct numbers being set in the register, the corresponding LED on the 7i31 LED Debug board varies in intensity as expected, and the small DC servo (driven by a 7i30 servo amp) changes its speed as expected. I'm not sure why you're seeing what you're seeing. Is maybe your pwmgen.scale not set to 1.0? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Field Programable Gate Array. is a programmable chip. Best Regards Kim Mortensen 2008/9/22 Eric H. Johnson [EMAIL PROTECTED] Sebastian, I am not sure what you mean by the FPGA register. Do you mean the interface register between the driver and the m5i20, or the register actually handling the PWM within the FPGA? As I said in an earlier email (off list), I concur that the value being sent to the FPGA through the interface register is correct (providing the bit shift to bits 16-27 is where the value is supposed to go), which would confirm that my pwmgen.scale is set to 1.0. I was able to measure the duty cycle on a pain old multi-meter. At values of 0.125 and below, the voltage would start dropping from the almost 100% duty cycle voltage of just under 3.3V, and very linearly for the range of 0.125 down to 0.0. The voltage reading from my meter was in fact very accurate. Here are the voltage readings I got: ValueVoltage =0.126 3.278 0.125 3.272 0.120 3.144 0.110 2.884 0.100 2.623 0.090 2.362 0.080 2.102 0.070 1.841 0.060 1.580 0.050 1.320 0.040 1.059 0.030 0.798 0.020 0.538 0.010 0.278 Regards, Eric How are you measuring duty cycle? I'm confident that (at least in my test setup) setting pwmgen.XX.value causes the correct number to be set in the FPGA register. Take a look at http://pastebin.ca/1208030. I updated pwmgen.c to decode the number in the PWM Value Register, then used raw.dump-state to show the register when pwmgen.value was -1.0 to +0.3 in increments of 0.1. Lines 1-22 show the relevant dmesg output for .value=0, then I just show the PWM Value Register for the other .value settings. In addition to the correct numbers being set in the register, the corresponding LED on the 7i31 LED Debug board varies in intensity as expected, and the small DC servo (driven by a 7i30 servo amp) changes its speed as expected. I'm not sure why you're seeing what you're seeing. Is maybe your pwmgen.scale not set to 1.0? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users - 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=100url=/___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Eric H. Johnson wrote: I am not sure what you mean by the FPGA register. Do you mean the interface register between the driver and the m5i20, or the register actually handling the PWM within the FPGA? As I said in an earlier email (off list), I concur that the value being sent to the FPGA through the interface register is correct (providing the bit shift to bits 16-27 is where the value is supposed to go), which would confirm that my pwmgen.scale is set to 1.0. By FPGA register i mean the register on the 5i20 board where the driver on the PC tells the 5i20 what PWM duty cycle it wants. I think this means both of the things you described. Ok so we're in agreement that the proper PWM value is being requested of the 5i20 board, and the question now is whether the FPGA outputs the proper PWM signal. I was able to measure the duty cycle on a pain old multi-meter. At values of 0.125 and below, the voltage would start dropping from the almost 100% duty cycle voltage of just under 3.3V, and very linearly for the range of 0.125 down to 0.0. The voltage reading from my meter was in fact very accurate. It is my understanding that a normal voltage meter is not a reliable tool for measuring the pulse width of a PWM signal. I'll see if I can borrow an oscilloscope from work... -- Sebastian Kuzminsky Theo: Julian? I haven't seen you in twenty years. You look good. The picture the police have of you doesn't do you justice. Julian: What do the police know about justice? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
On Mon, 22 Sep 2008, Sebastian Kuzminsky wrote: Date: Mon, 22 Sep 2008 07:26:19 -0600 From: Sebastian Kuzminsky [EMAIL PROTECTED] Reply-To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net Subject: Re: [Emc-users] stepgen pwmgen fixes Eric H. Johnson wrote: I am not sure what you mean by the FPGA register. Do you mean the interface register between the driver and the m5i20, or the register actually handling the PWM within the FPGA? As I said in an earlier email (off list), I concur that the value being sent to the FPGA through the interface register is correct (providing the bit shift to bits 16-27 is where the value is supposed to go), which would confirm that my pwmgen.scale is set to 1.0. By FPGA register i mean the register on the 5i20 board where the driver on the PC tells the 5i20 what PWM duty cycle it wants. I think this means both of the things you described. Ok so we're in agreement that the proper PWM value is being requested of the 5i20 board, and the question now is whether the FPGA outputs the proper PWM signal. I was able to measure the duty cycle on a pain old multi-meter. At values of 0.125 and below, the voltage would start dropping from the almost 100% duty cycle voltage of just under 3.3V, and very linearly for the range of 0.125 down to 0.0. The voltage reading from my meter was in fact very accurate. It is my understanding that a normal voltage meter is not a reliable tool for measuring the pulse width of a PWM signal. I'll see if I can borrow an oscilloscope from work... My experience is that most DVMs and even analog VOMs tend to read PWM average pretty well when set to read DC volts. This is because DVM DC input circuits usually include a low pass filter that does a good job of averaging the PWM and analog VOMs meter mechanics do the same thing. The symptoms sound like you are running in 9 bit PWM mode (but the driver expecting 12 bit PWM behaviour), that is the PWM width register ends up being 00b instead of the desired 11b. This could be because the driver does not initialize it or manages to overwrite the width field in the mode register after its been set. It could also be a FPGA firmware error, but I checked the HostMot2 firmware fairy thoroughly this February, I can do some low level poking and check this today. (Looking at my coffee cup that says Don't look at me, Its clearly a software problem) -- Sebastian Kuzminsky Theo: Julian? I haven't seen you in twenty years. You look good. The picture the police have of you doesn't do you justice. Julian: What do the police know about justice? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users Peter Wallace Mesa Electronics (\__/) (='.'=) This is Bunny. Copy and paste bunny into your ()_() signature to help him gain world domination. - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Peter C. Wallace wrote: The symptoms sound like you are running in 9 bit PWM mode (but the driver expecting 12 bit PWM behaviour), that is the PWM width register ends up being 00b instead of the desired 11b. This could be because the driver does not initialize it or manages to overwrite the width field in the mode register after its been set. Eric, could you set your machine up in the way it is when PWM is misbehaving for you, the setp hm2_5i20.0.raw.dump-state 1, then pastebin the relevant part of the dmesg? That'll show (among other useful things) what pwm width the hm2 driver thinks it's running in. FWIW, the dmesg info i pastebinned last night (http://pastebin.ca/1208030) shows the PWM Mode register set to 0x23, which means 12-bit PWM, sawtooth (not symmetrical), normal sign/magnitude on the normal pins, double buffer mode. I'm curious if yours is somehow set differently. -- Sebastian Kuzminsky Theo: Julian? I haven't seen you in twenty years. You look good. The picture the police have of you doesn't do you justice. Julian: What do the police know about justice? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
On Mon, 22 Sep 2008, Sebastian Kuzminsky wrote: Date: Mon, 22 Sep 2008 09:52:09 -0600 From: Sebastian Kuzminsky [EMAIL PROTECTED] Reply-To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net Subject: Re: [Emc-users] stepgen pwmgen fixes Eric H. Johnson wrote: I realize that a voltmeter may not be a reliable means of measuring a PWM, but my particular voltmeter does seem to agree with the power output of the laser. It also agrees with the voltages I read when using the standard m5i20 driver, which is already known to produce good values. Thus, given the consistency and linearity of the readings, I am fairly confident that they reasonably reflect what is going on with the PWM output. Peter, do you have any input on this? I think an oscilloscope trace of the PWM pin being driven by the hostmot2 driver is needed to move forward on this issue. SVST8_4.BIT (2/01/08) MD5 (SVST8_4.BIT) = dcf2cabf2a4b3fc08d8b783bbaaddeab Heres my little test batch file: wpo 1100 00fc0fc0 ; enable outputs wpo 1200 00fc0fc0 ; select alternate sources wpo 4200 ; PWM rate max wpo 4400 : turn on enables wpo 4100 20 ; 9 bit pwm mode wpo 4000 0100 ; 1/2 way up = 256 pause ok = 50% duty cycle at 195.31 KHz wpo 4100 21 ; 10 bit PWM mode wpo 4000 0200 ; 1/2 way up = 512 pause ok = 50% duty cycle at 97.66 KHz wpo 4100 22 ; 11 bit PWM mode wpo 4000 0400 ; 1/2 way up = 1024 pause ok = 50% duty cycle at 48.83 Khz wpo 4100 23 ; 12 bit PWM mode wpo 4000 0800 ; 1/2 way up = 2048 pause ok = 50% duty cycle at 24.41 Khz All looks to be OK Peter Wallace Mesa Electronics (\__/) (='.'=) This is Bunny. Copy and paste bunny into your ()_() signature to help him gain world domination. - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Peter C. Wallace wrote: On Mon, 22 Sep 2008, Sebastian Kuzminsky wrote: Eric H. Johnson wrote: I realize that a voltmeter may not be a reliable means of measuring a PWM, but my particular voltmeter does seem to agree with the power output of the laser. It also agrees with the voltages I read when using the standard m5i20 driver, which is already known to produce good values. Thus, given the consistency and linearity of the readings, I am fairly confident that they reasonably reflect what is going on with the PWM output. Peter, do you have any input on this? I think an oscilloscope trace of the PWM pin being driven by the hostmot2 driver is needed to move forward on this issue. SVST8_4.BIT (2/01/08) MD5 (SVST8_4.BIT) = dcf2cabf2a4b3fc08d8b783bbaaddeab This is the same md5sum as the mesa-hostmot2/firmware/5i20/SVST8_4.BIT in our CVS tree. Heres my little test batch file: wpo 1100 00fc0fc0 ; enable outputs wpo 1200 00fc0fc0 ; select alternate sources wpo 4200 ; PWM rate max wpo 4400 : turn on enables wpo 4100 20 ; 9 bit pwm mode wpo 4000 0100 ; 1/2 way up = 256 pause ok = 50% duty cycle at 195.31 KHz wpo 4100 21 ; 10 bit PWM mode wpo 4000 0200 ; 1/2 way up = 512 pause ok = 50% duty cycle at 97.66 KHz wpo 4100 22 ; 11 bit PWM mode wpo 4000 0400 ; 1/2 way up = 1024 pause ok = 50% duty cycle at 48.83 Khz wpo 4100 23 ; 12 bit PWM mode wpo 4000 0800 ; 1/2 way up = 2048 pause ok = 50% duty cycle at 24.41 Khz All looks to be OK Great to hear. Peter's test does raw I/O with the 5i20 to cause it to output the PWM signal, so there's still a possibility that the hostmot2 driver does a different pattern of I/O which results in an incorrect waveform output. I've found a scope I can borrow, I'll do a similar test using the hostmot2 driver tonight or another night later this week. Does anyone know what the PWM frequency of the old m5i20 driver is? -- Sebastian Kuzminsky Theo: Julian? I haven't seen you in twenty years. You look good. The picture the police have of you doesn't do you justice. Julian: What do the police know about justice? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Sebastian, It is specified as a parameter to the driver, up to 33Khz IIRC. I was setting it to 5Khz. Regards, Eric Does anyone know what the PWM frequency of the old m5i20 driver is? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Eric H. Johnson wrote: It is specified as a parameter to the driver, up to 33Khz IIRC. I was setting it to 5Khz. I bet that's it then. The hm2_5i20 pwmgen currently goes at 25 KHz with no way to change it. I'll add a HAL parameter and let's see if that helps. -- Sebastian Kuzminsky Theo: Julian? I haven't seen you in twenty years. You look good. The picture the police have of you doesn't do you justice. Julian: What do the police know about justice? - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
Eric (and anyone else who's had problems with pwmgen), please try this: http://highlab.com/~seb/bzr/emc2/hostmot2/src/hal/drivers/mesa-hostmot2/pwmgen.c Lucky for me not many people use bzr, or you could easily see how stupid my bug was. I think I'm safe from the humiliation until I commit it to the CVS tree later tonight... -- Sebastian Kuzminsky On the molecular scale, you find it's reasonable to have a machine that does a million steps per second, a mechanical system that works at computer speeds. -- K. Eric Drexler - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] stepgen pwmgen fixes
On Sunday 21 September 2008 16:04:48 Eric Johnson wrote: As for the PWM, I think I have something for you. It looks to me like something in the scaling must not be working in FPGA. Fortunately it does appear to be linear, so I think I can get my system going by adjusting my scaling. It may affect the granularity, but for this application that really isn't an issue. Here is what I am getting; Any value for the PWM set at 0.126 and above results in a duty cycle of almost 100% (99.3). At 0.125 the duty cycle is 99.1. The following table lists the duty cycles for values of 0.120 to 0.010 in increments of 0.010. ValueDuty Cycle 0.120 95.3 0.110 87.3 0.100 79.5 0.090 71.6 0.080 63.7 0.070 55.8 0.060 47.9 0.050 40.0 0.040 32.1 0.030 24.2 0.020 16.3 0.010 8.4 Pretty consistent at 7.9% per 0.010 of value. As a hunch, it looks to me like you are off by 4 bits some place, because 0.125 is 1/8th of the range of 0.0 to 1.0. How are you measuring duty cycle? I'm confident that (at least in my test setup) setting pwmgen.XX.value causes the correct number to be set in the FPGA register. Take a look at http://pastebin.ca/1208030. I updated pwmgen.c to decode the number in the PWM Value Register, then used raw.dump-state to show the register when pwmgen.value was -1.0 to +0.3 in increments of 0.1. Lines 1-22 show the relevant dmesg output for .value=0, then I just show the PWM Value Register for the other .value settings. In addition to the correct numbers being set in the register, the corresponding LED on the 7i31 LED Debug board varies in intensity as expected, and the small DC servo (driven by a 7i30 servo amp) changes its speed as expected. I'm not sure why you're seeing what you're seeing. Is maybe your pwmgen.scale not set to 1.0? -- Sebastian Kuzminsky On the molecular scale, you find it's reasonable to have a machine that does a million steps per second, a mechanical system that works at computer speeds. -- K. Eric Drexler - 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=100url=/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
