[Sbs_gems] [EXTERNAL] Wiener HV modules not recommended for parallel-use.

[Liyanage, Nilanga K (nl8n)]

HI John

Thanks, this is very useful information. So looks like connecting supplies in parallel is not a good option.

Another possible way is to use one channel to supply HV to one electrode in a GEM. For example one channel set to provide voltage to bottom GEM bottom side, next channel to bottom GEM top side and so on. The channels will have different high voltages, but all referenced to the common ground.

But the question is about tripping the channels; for safety, we want all channels to trip, if one channel trips.

I would appreciate if you could check with Wiener folks to see if the channels could be coupled so that they all trip together.

Best

Nilanga
________________________________
From: Sbs_gems <sbs_gems-bounces at jlab.org> on behalf of John Boyd <jab7bp at virginia.edu>
Sent: Thursday, February 17, 2022 4:02 PM
To: Sbs_gems at jlab.org <sbs_gems at jlab.org>
Subject: [Sbs_gems] [EXTERNAL] Wiener HV modules not recommended for parallel-use.

I spoke with the engineers at Wiener today and discussed with them our ideas for running a higher current (greater than 1 mA) configuration by running channels/modules in parallel. Both engineers, Bryce and Andreas, are familiar with our application -- using these for GEMs -- and how everything is meant to supply and work together (they've worked pretty extensively with me/us for a few years).

The final conclusion of my conversation with them was that it is not recommended to run these supplies in parallel to try and get the two 1 mA supplies to feed more than 1 mA. They even sent this question back to the main facility in Germany with the same recommendation.

The concern they had was that there would probably be inconsistencies or problematic issues with the regulation of the voltage or current. If the supply experienced any oscillation in trying to maintain voltage/current that was not properly regulated then this oscillation could cause serious problems to either our devices (GEMs) or the HV modules themselves.

Another issue would be the modules' ability to ensure that the balancing between the two parallel "legs" is consistent. Without proper redundant feedback loops and reverse polarity diodes feeding back into the main control module (so additional modifications would be needed to get this interfacing at inside the module) it would be difficult to make sure that both "legs" were trying to occupy an equal portion of the potential (if one side tried to supply more than it may oscillate into a much higher range and then reach a current limit thereby tripping and leaving only the other leg still energized and feeding the single leg).

I hope this all makes sense. I gathered this all through conversation so hopefully I interpreted and relayed this information correctly.

They did offer to provide us with loaner (old version) HV power supplies that we could use to test our idea.

The lead time for any of their modules is 12+ weeks.

They do have a HV module that would work (4kV at 2mA) but it has the lead time stated above.

Thank you and please offer any comments/concerns.

Best,

John
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/sbs_gems/attachments/20220218/84f828cf/attachment.html>