Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: GregMill on August 01, 2007, 02:36:31 PM

Title: Clarification of charge pump frequency
Post by: GregMill on August 01, 2007, 02:36:31 PM
I have been having problems with my charge pump circuit working correctly and I have narrowed it down to most likely be a charge pump hardware issue on my breakout board. While debugging the problem I measured the charge pump frequency to be ~6.4kHz with a period of 156us using an Oscilloscope. This appears to be half the listed frequency of 12.5kHz given in the documentation. This was very confusing to me and I thought I had a serious driver problem. I measured the same frequency on 2 other PCs and the driver test runs at a very smooth ~25kHz rate. So the conclusion I have drawn is that the 12.5kHz listed rate is not the pulse frequency (measured rising edge to rising edge with 50% duty cycle) but rather the update rate of the pulse generator (rising edge to falling edge to rising edge…) resulting in the pulse frequency being half the update rate frequency. Is my assumption correct or am I making a newbie error in blindness or misunderstanding.

Thanks for you clarification,
Greg
Title: Re: Clarification of charge pump frequency
Post by: Whacko on August 01, 2007, 06:30:52 PM
I haven't scoped mine, but I run the signal through a MCP1407 and a pulse transformer, so if there's no frequency, there can be no transferred signal via the pulse trafo. Works very well. Are you having a problem with the low frequency? Is the rest of your hardware/software working?

Regards,
William
Title: Re: Clarification of charge pump frequency
Post by: GregMill on August 01, 2007, 07:33:14 PM
Everything else is working perfectly.
Mach3 is the best!!!!.
I am working with the board vendor to find a solution to the problem with the charge pump circuit. If I have to I am going to use a microcontroller to measure the charge pump frequency and output a 'charge pump is good' signal as this will be the most robust and more importantly I have one I am comfortable programming sitting right in front of me  ;)

Thanks,
Greg
Title: Re: Clarification of charge pump frequency
Post by: Jeff_Birt on August 01, 2007, 10:10:26 PM
The frequency is 12.5 kHz.  There is a setting, somewhere, in mach that changes it to 7 kHz as that is what the samll laser cutter/engravers use.  My guess is that you turned that setting on by mistake.
Title: Re: Clarification of charge pump frequency
Post by: coyotegd on September 07, 2007, 07:06:11 PM
I have been working on a breakout board with a microcontroller monitor.  Every now and then when I start Mach3, it serves up a 6.4kHz chargepump signal.  If I close Mach3 and restart, 12.1kHz.  There's a bug somewhere.
Title: Re: Clarification of charge pump frequency
Post by: PRT on January 21, 2008, 09:07:45 AM
I found a setting in Mach's General config - "Set charge pump to 5KHz for laser standy level"
maybe you had a problem with saving your settings...
Title: Re: Clarification of charge pump frequency
Post by: TonyP on January 21, 2008, 11:52:05 AM
Greg
Are you actually monitoring the charge pump signal on the port pin with the unit plugged in or after any opto- isolator? I had a problem when I was designing a new breakout board. The LPT1 port I was using would not drive very much at all. I sorted it out by  using  buffers before the optos. Many pump circuits are fairly tolerent of absolute frequency - its really missing pulses that should shut them down.

Tony
Title: Re: Clarification of charge pump frequency
Post by: TonyP on January 21, 2008, 12:22:21 PM
Another thing. I've just run up a spare board & and measured the frequency. I found mine was running at 2kHz. So I upped the kernel speed &, sure enough, the frequency went to around 12kHz with a speed of 60000. So it looks as though the frequency isn't fixed anyway. My charge pump still operated when the software came up though.

Tony
Title: Re: Clarification of charge pump frequency
Post by: PRT on January 22, 2008, 02:06:24 AM
We use opto-isolation straight after the port pin, and get a result of ~2.4KHz at every kernel speed. Might be we used wrong opto-isolators, or a wrong transistor (for converting to 24VDC), as I test the freq. on a PLC. Still a couple tests to go.

Wojtek