horiba-ihr320: Commands are only partially executed

[flowbee74]

Hi. I am trying run client on a windows 10 machine.
yaqd-horiba-ihr320 stats up with some "minor errors"

<3> ERR : 2023-05-16T14:47:56+0200 : mono : USBTimeoutError(10060, 'Operation timed out')

and sometimes ceonnection are lost
<6> INFO : 2023-05-16T14:49:59+0200 : mono : Connection made from ('127.0.0.1', 57396) to mono
<6> INFO : 2023-05-16T15:04:32+0200 : mono : Connection lost from ('127.0.0.1', 57396) to mono
<6> INFO : 2023-05-16T15:44:43+0200 : mono : Connection made from ('127.0.0.1', 57602) to mono

otherwise I can read out values (slit witdh, mirror position etc.) easily. The problem is that some commands are working flawlessly and some are only partially executed. For instance,

Changing the grating position works perfectly
Cahnging slit width gets stuck after moving them a little
Flipping the mirrors works super slowly - if at all.

Any ideas ?

[DavidRivas3146]

Hi,
amazing project!
Just a question, I also experienced some problems when changing the slit size. Basically, when I enter a slit size value, sometimes it does not correspond to what is observed on the iHR320 spectrometer. Also, sometimes the slit gets stuck. Is there any possible solution? thanks!

[idanshv]

Hi, @flowbee74 @DavidRivas3146

I’m experiencing the same issues with the Horiba-ihr320. Specifically, the commands related to changing and retrieving the width of the slits are not working properly.
The commands to get the slit width are returning inconsistent and incorrect values that do not correspond to the actual slit width limits.
Did you manage to diagnose or resolve these issues? Any help or suggestions would be greatly appreciated!

Thanks!

[ksunden]

Unfortunately, I no longer have regular access to the hardware for this program, so changes are likely to be slow (if at all).

As to the unreliability, I saw some of that when I used this regularly, but it was honestly not that impactful to my usage (i.e. it did not affect the most common actions for e.g. scanning the mono) and it was never often enough for me to be able to pin anything down... See also #19, which is an attempt, at least during the startup procedure, to improve reliability. My main hangup there is that I do not believe it has been tested on hardware, with my secondary hangup being unbounded looping (testing on hardware could allow us to set limits so that 99.9+% of the time it succeeds, but in that rare instance it doesn't it doesn't just keep trying ad infinitum, but not sure how long is "long enough" without trying).

As for the slit width, @idanshv, there are two different slit widths available, and I made this with the one I had in mind. There are a 7 mm version and a 2 mm version, and the one I had was 7 mm. The actual numbers that are sent on the wire are integers 0-1000, as a fraction of the max slit width, controlled by this value:

yaqd-horiba/yaqd_horiba/_horiba_ihr.py

Line 53 in 6e39599

const = 7 / 1000

If you have a 2 mm slit, it will be off by a factor of 7/2.
I am fully willing to make that value configurable, just never did because I only knew of one instance using it when I originally wrote this.