Hi! >> http://www.produktinfo.conrad.com/datenblaetter/950000-974999/967599-an-01-en-SMC_800_Schrittmotor_Karte.pdf >> >> Signals are described on page 11. > There has to be something wrong with that information, as it gives the > same bit pattern for 60% and 100% current. Yes, of course. It is an error in the english manual. The german one is bug-free ;).
> > That can _almost_ be controlled by the Type 15 stepgen which lets you > supply a lookup table of your own. It ought to be fairly simple to > modify it to create 6 pins. (changes in lines stepgen.c 1211 and 1217 > to add more phases, 308 to add more steps, and you can recompile just > that file with comp (I think). Hmm, I've thought about that too. To me it was confusing, that the number of phases was maxed to 5. Is the number of phases limited only by the output pins or are there any other (timing) constraints? > However, the fact that you can only step one axis at a time looks a > lot more troublesome. I _might_ be possible to enable three stepgens > one at a time, and direct their output to the parport accordingly with > a bunch of AND gates (in HAL, i.e. software). That is probably easier > than writing a step generator component from scratch. Ok, so let's assume I can modify stepgen.c - How is the actual stepping done? Is it one step every base-period thread or is it running asynchronous to the base thread and only reports back when the motion-cmd is finished? And does stepgen have a feedback pin like x-step-done (Which then could be used for the HAL logic)? I agree that the logic to deliver the x-y-z sequence could be a painful task. After all, I think I will spend another two or three days to check if I can do it in HAL after I set up my Milling PC, if not the boards from http://www.pololu.com/catalog/product/1182 really look like a good buy ;-) Am 08.08.2012 11:44, schrieb andy pugh: > On 8 August 2012 06:50, Peter Blodow <p.blo...@dreki.de> wrote: >> Gentlemen, >> I am running the big brother of this card named SMC-1500 including a >> small piggy back board with some intelligence on it. > That piggy-back board makes the drives into standard step/dir drives, > and allows the computer to control the axes independently. > Without that PCB the computer needs to select which axis to control, > then send the phase pattern data. > > It would be relatively simple to link the two with something like an > Arduino, but needing 15 outputs and 6 inputs I think it would need to > be a Mega, and cost about the same as a complete new driver board. > hmm, I have some uC-Boards in my personal stock, maybe just writing a small piece of code to make a similar interface board like SMC1500 has would be another alternative. Am 08.08.2012 12:59, schrieb Peter Blodow: > Andy, > of course, that board design is about 25 to 30 years old. I have several > of them since Olim's time, i.e., DOS 3.3 ages. The SMC 800 (as well as > the naked SMC1500) can be run with the drivers that EMIS delivered with > the board. At those times, every user had to write his own stepper > software, and a ready made driver including trajectory planning was > highly welcome. The C and PASCAL drivers include the features of ramping > and drive the 3 axes simultaneously, also on the SMC 800! The piggy > back board on the SMC 1500 is only needed when you are not using these > drivers and write your own software. The drivers can still be downloaded > from EMIS as well as MS-Windows software. > > As the boards are apparently undestroyable, they still are in use around > the country and they still sell them today although the design is not > quite up to date. I once killed a transistor array by thermics and had > to replace it, that was all. > > Peter Blodow I took a look at the DOS driver, looks like it drives the board in half-step mode in a x-y-z sequence that is dependent on the input vector. They are not driven simultaneously, they cannot be. But to exchange the outb - commands to HAL pins should do the trick here. I am thinking of using this driver like another stepgen. Maybe someone can give me a hint how the movement commands look like (are there X-Y-Z vectors available) If such vectors could be generated somehow, the second task would be to sync timings. That sounds like an alternative to me..... Thanks for all your hints, I will keep you up to date as my work continues ;-) Alexander ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
