Ethercat HAL driver

Thanks Again guys
roschi now the numbers start making sense I will have a go at it. Do I still need to use the hall complex to specify the scale? If not how do I scale it? it is very frustrating knowing that it is only going to be a little thing that is stopping you in your tracks, but I get so focused that I can't see it! Anyway, this will make me learn more, and I sent you an email about this problem. But I also wanted to let you know about those lunyee drives with the
second encoder feedback loop from linear encoder etc. and your thoughts on them they are SV-DA200
Anyway, I will have a go at it tomorrow.

Ok I have removed the halType complex and if I put the scale to "1" i can get them operational but as you could imagine with the scale so low that half a turn shows as 500mm of travel on the axis DRO. now it I put scale to 10 I can start and get operational but as soon as I go to jog an axis it moves a little bit and throws up a joint following error for that axis also shows very large numbers on the axis DRO.
if I put the scale to 100 the drives throw an A07 alarm again

So I had to walk away from this for a while and think about what is happening. I came back tried a few things to no avail but
I have noticed when i put any number larger than 1 I get the joint folowing error. If I try to toggle machine power agian the sevo trys to run flat out for les then a second and throws a joint error. And the DRO numbers coming up for that axis are obsinely large like upto a km
(would be a big mill!) on the last setup scale="1000" and it worked flawlessly now is the issue to do with the halType special32 and im now using halType float? So I am unsure how the scale is working i have looked through the code on the git repository and tried to see how the scale system works and all i can say is i need to learn a lot more about some of the concepts in cpp but i can see how it is used in roschi's halType="special32" but i cant quite get a grip of how it works with the halType float I thin I might have to implement The special32 I can already feel the grey hair (well greyer hair)

My CanOpen manual says:
TARGET POSITION INDEX 0X607A
Type Access Units Range Map PDO Memory
Integer 32 RW counts - YES RF

Why should your 0X607A be a float?

Hi sirop
I was only using float so that I could add scale. the generic.c will check that you have specified float if you have scale
as part of its error handling. I will spend a bit more time tonight going over the code to see how the scaling is implemented and I am looking to see if there is some sort of packet sniffer so I can see what is in the EtherCAT packet.
but this morning I have to drive into town its about 135km each way!

Sorry at all a little info ... is possible to run ethercat driver over linuxmint 18 rt_preempt?

regards
BKT