On Friday 26 February 2021 16:37:02 Peter C. Wallace wrote: > >The 7i90 is designed to be programmed from an epp parport, and I > > don't believe it can be programmed again once an spi interface has > > been put in it with out doing some jumper moving to reset it to > > parport drive before it can be reprogrammed again. Its explained in > > the docs for it. It is also an unbuffered 3.3 volt device, and any > > noise below ground or above 3.4 volts will destroy it, so make sure > > you have enough 7i42TA's to buffer and protect it. A single bolt > > star ground is very important around the 7i90. Each 7i42TA buffers 1 > > 50 pin, 24 i/o lines, so all 72 needs 3 of them and short 50 pin to > > 50 pin ribbon jumpers. Substitute 7i33TA's for the 7i42TA's in your > > case so you'll need two 7i33TA's and one 7i42TA for the gpio stuffs > > since each 7i33TA only handles 4 servo's including their encoder > > returns. The TA versions are worth their price in bottled beer > > because of the built in screw terminals. > > > >Cheers, Gene Heskett > > Thats not quite true, the 7I90 I/O pins will all take voltages from > about -0.5V to +7V. That said, the I/O pins are bare IC pins and are > easy to damage with negative inputs unless there is some current > limiting. This is because all I/O pins have ESD protection diodes with > anode to ground and cathode to the I/O pin. These diodes are tiny and > can be damaged by current > > > 10 mA. How do you get a negative input? Its very easy, simply > > connecting > > an I/O pin to frame ground will do it if there is any high frequency, > high current device in the immediate area (Step drives, servo drives, > VFDs etc) Also connecting any external device to the FPGA card when a > common ground has not been established between the device and the FPGA > card will do it as well. If you are using bare FPGA pins and you have > noisy grounds you can protect the I/O pins by adding a series resistor > (say 470 Ohms)
My initial wire-up didn't have a good star ground, Peter. I blew the first 2 7i90's quickly without the 7i42TA's. On hooking up my then best scope I was blown away by finding the noise levels I observed on a Hitachi v1065 scope, a 100 MHz dual trace. Up to 80 volts p-p with nearly 100 MHz rise and fall times. But it was analog, and quite dim at that writing speed. So I made myself a birthday present 5 years ago of a new digital sampler which was able to capture that stuff much brighter, with its 1GHz sampler head. I could actually see those diodes turning on. Very educational. So with the passing of the woof, I have now obtained Siglent's best 10" 4 trace, with a 350 MHz bandwidth from a 2GHZ sampler. A truly amazing piece of test gear. So are the 10x probes, rated at 350 MHz, yet they are nearly 6 feet long. Noise does not now stand a chance of being undetected. But with a common ground bolt, its under a half volt now. > Peter Wallace > Mesa Electronics > > > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/gene> _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
