High purity Germanium detector design related clarifications

This area is for discussions involving any fusion related radiation metrology issues. Neutrons are the key signature of fusion, but other radiations are of interest to the amateur fusioneer as well.
AshishGopalakrishnan
Posts: 8
Joined: Wed Aug 23, 2017 10:57 pm
Real name: Ashish Gopalakrishnan

High purity Germanium detector design related clarifications

Post by AshishGopalakrishnan » Wed Aug 23, 2017 11:10 pm

Hello everyone,

I am trying to do a case study on some of the design related flaws in a high purity germanium detector. So there is a cylindrical germanium crystal (weighing about a kilogram) that is connected to a vertical copper rod. The setup is shown in the figure that I have attached with this post.

There is a clear triangular plate in this setup that I have marked in yellow. This plate cracks even with the most gentlest handling. So I would like to get your ideas on the following:
1) Which one is optimum for the triangular plate: high thermal conductivity or low thermal conductivity?
2) Since it cracks, what material could it have been made?
3) Why would one have chosen this geometry and material? Is this plate geometry a good design?

The crystal must be kept as cold as possible. Hence, the entire setup is connected to a 30 L liquid nitrogen dewar while running. In addition, the detector is under high vacuum to minimize convective heat transfer between the room-temperature
exterior of the detector and the interior cold crystal and copper rods.

So, for the above questions, I feel that as cold environment is more suitable for operation, the plate should have low thermal conductivity, as even the slightest rise in temperature will reduce the efficiency of operation. I am pretty unsure of the material and the reasoning behind this design.

Any suggestions and ideas that would give more insights into the above questions will be very helpful.


Thanks,
Ashish
Attachments
Capture.JPG

User avatar
Richard Hull
Site Admin
Posts: 10899
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: High purity Germanium detector design related clarificat

Post by Richard Hull » Thu Aug 24, 2017 6:22 am

The triangular plate is for insulation and support which must be superior due to the fact that a form of highspeed detection using ultra-high impedance must be maintained as the current pulses from the typical specially selected FET are miniscule.

If it cracks, did you crack it via over tightening of fittings? Plexiglass, cracks real easy under stress tensioning. I am not sure, but once under LN2, the characteristics of the plastic may change forever, especially if brought back up to room temperature. I assume it is a simple triangle of plexiglass. It looks to be 1/4-inch thick!

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

AshishGopalakrishnan
Posts: 8
Joined: Wed Aug 23, 2017 10:57 pm
Real name: Ashish Gopalakrishnan

Re: High purity Germanium detector design related clarificat

Post by AshishGopalakrishnan » Thu Aug 24, 2017 7:42 am

Hi Richard,

Thank you for your response. I also feel that the screws could have cracked the plexiglass.
Richard Hull wrote:
If it cracks, did you crack it via over tightening of fittings? Plexiglass, cracks real easy under stress tensioning. I am not sure, but once under LN2, the characteristics of the plastic may change forever, especially if brought back up to room temperature. I assume it is a simple triangle of plexiglass. It looks to be 1/4-inch thick!

Richard Hull
This is one of the existing designs of a detector, which is currently not being sold in the market. I got this information and the pic from the manufacturer. I will have to submit my ideas about the possible flaws in this design and also suggest possible improvements. Hence, I was trying to understand the reasoning behind such a design for the plate. I understand that there has to be some form of insulation since the germanium crystal has to be maintained at a cold temperature during operation and any form of temperature rise will decrease the efficiency. Hence, one would have chosen Plexiglass for its low thermal conductivity.

However, I am still confused about the design of the triangular plate. I feel that the screws also constrain its movement and as I have mentioned in the previous post, even the gentlest handling causes cracks, which could be due to the fact that the motion of the copper rod in any direction is constrained by the plate. Please let me know your suggestions on my insights.

It would also be helpful if you can give your ideas on what could be the design flaw and possible ways to fix it in terms of geometry and material.

Thanks
Ashish

John Futter
Posts: 1296
Joined: Thu Apr 22, 2004 2:29 am
Real name:
Contact:

Re: High purity Germanium detector design related clarificat

Post by John Futter » Thu Aug 24, 2017 8:33 am

I do not know what happened to my last reply
Acryllic plate (perspex) is used extensively in nuclear physics
It is not a perfect insulator it has a known Gohms per sq but only for cast acryllic
extruded has stress built in which alters the Gohm per sq in proportion to stress
Having a known Gohm per sq allows E field compensation.
dont forget one end of that crystal is anything up to 3kV of reverse bias creating the depletion layer (active layer)
we have many high purity Ge detectors and we try to keep them @ LN2 temp all the time
every thermal cycle of LN2 to room temp and back to LN2 temp causes stress in every component especially the low noise fet and associated electronics in the head of that unit
Our oldest Ge is 40 years old and going strong -- it may come up to room temp every couple of years due to error but probably has done less than 20 thermal cycles.
I have heard that annealing extruded acyllic getting it near melting relieves the stress it makes sense but I have no experience with doing this.
The clamping force and clearance of the holes through the acyllic is important get it wrong and it will craze or crack

User avatar
Finn Hammer
Posts: 84
Joined: Sat Mar 05, 2016 12:21 pm
Real name: Finn Hammer
Contact:

Re: High purity Germanium detector design related clarificat

Post by Finn Hammer » Thu Aug 24, 2017 8:45 am

gkashish, (You really should introduce yourself first, and also use your real name as handle)

A 3 point connection between standoffs and a plate is a common engineering practice, because you get the possibility of a fairly stress free connection without too strict restrictions on dimensional tolerance on the length of the standoffs. Think of a 3 legged stool, or a tripod, it is always stable on the ground. With this in mind, 5 or 6 standoffs would make a better support.
However, the very large holes between standoff and center of the plate, where the wires pass trough,
Udklip.JPG
Red lines shows the big holes (feed through holes) in the supporting plate
are definitely not making things any better, there is not much material left to support a structure weighing in at kilograms, particularly not when this weight is waving at the end of a stick
All this said, you don't stand a fair chance saying much of real value, as long as the wording is along the lines of "even with the most gentlest handling", nor that you have a picture showing exactly where the crack occurs.

My response would be something like this:

1. Make the plate round, and save the machining operation where it is made triangular
2. Rotate the hole pattern 60 degrees between feed through holes an fastener holes
3. Reduce size of feed through holes
4. use more standoffs.

This will make the design a whole lot sturdier.

Edit:
I just read John's response which reminds me of my days at Radiometer in Copenhagen, where I was method technician in mechanical engineering section.
We made several heavily machined parts out of acryllic, and these parts had to be annealed between roughing and finishing passes in the lathe, sometimes several times, depending on the intricacy of the part. I don't know the temperatures involved, I just specified it. But it is important to do.

Cheers, Finn Hammer

User avatar
Dennis P Brown
Posts: 1523
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: High purity Germanium detector design related clarificat

Post by Dennis P Brown » Thu Aug 24, 2017 10:31 am

As cold as possible using just liquid nitrogen!? Really? One could use a closed circuit He refrigerator that could get to a few Kelvin at only a small cost in electrical usage (we have two such units used to cool detectors) - obviously, we have little details to go on for why that detector has to use of LN2 (optimum temp, maybe?) While closed He based cooling units can't handle a large heat load,but since I am not aware of the design details of your Ge detector, such a unit might or might not be applicable.

One could use ceramic for both strength and resistance to cracking at a slight cost in thermal conduction - a trade off; considering how cheap LN2 is compared to that massive Ge crystal and the danger to that crystal if the mount fails, I would consider using a far stronger material like a ceramic for the support at the cost of a bit more LN2 use. Failing that, teflon can be used instead and offers better crack resistance (we use teflon (for a similar purpose but we get to under 3 K.)

John Futter
Posts: 1296
Joined: Thu Apr 22, 2004 2:29 am
Real name:
Contact:

Re: High purity Germanium detector design related clarificat

Post by John Futter » Thu Aug 24, 2017 7:51 pm

Dennis
When these things were designed there were no cryo coolers
the detectors also tend to be microphonic so a cryo cooler would have to be isolated

AshishGopalakrishnan
Posts: 8
Joined: Wed Aug 23, 2017 10:57 pm
Real name: Ashish Gopalakrishnan

Re: High purity Germanium detector design related clarificat

Post by AshishGopalakrishnan » Thu Aug 24, 2017 11:24 pm

Thank you for your responses John Futter, Finn Hammer and Dennis P Brown.

Finn, my apologies for a short introduction. I am a mechanical engineering graduate from Arizona State University. I am currently doing a case study on the finding the design flaws in the detector that is shown in the picture, which is one of the existing designs and currently not being sold in the market. I got this information and the pic from the manufacturer. I will have to submit my ideas about the possible flaws in this design and also suggest possible improvements.

Regarding the cracks, please find the attachment. The plate cracks in the mid-span and around the screws. Hence, from all your suggestions, I feel that the stresses due to thermal cycling and the ones from tightening the bolts could cause it to crack.

Another information I would like to share is that during manufacture, this detector assemble was baked at 100 C at vacuum to remove volatile contaminants like water and adsorbed gases.

As you advised, I could add your points for a better design.

However, is there any other material apart from Plexiglas for the crystal support that would satisfy the following conditions:
-provide thermal insulation
-withstand thermal cycling, vacuum conditions, and also baking temperature.


Thanks,
Ashish
Attachments
Image with cracks.JPG

User avatar
Dennis P Brown
Posts: 1523
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: High purity Germanium detector design related clarificat

Post by Dennis P Brown » Fri Aug 25, 2017 11:07 am

I assume you read all of my post? Teflon can meet those requirements you just reposted to replace the rigid plastic in that support bracket- while it does have dimensionality issues at room temp when machined; however, a simple solution to that problem is just use somewhat thicker sections. Teflon is vastly superior to those other plastics relative to its resistance to cracking - also, teflon has no issues at 100 C.

While ceramic isn't in the same class as plastics/teflon relative to lower thermal conduction, simple teflon washers and insulating (plastic) bolts can help mitigate that problem (maybe enough or not; that might require a quick calculation.) Machinable ceramics are vastly stronger, offer extreme dimension stability and far better resistance to cracking, bake out performance and vacuum stability characteristics than those rigid plastics (but certainly not as resistant to cracking compared to teflon.) One just needs to bother with better design to exploit this material.

Still, overall, teflon meets all your requirements and would be your best easy solution.

Aside: rather strange that such large holes are drilled through the plastic support - unless needed (and the photo shows there not), why do that? If the issue is thermal conduction than someone is terribly misled; at LN2 temps, increasing the size of those holes is essentially meaningless for thermal exchange issue across that bracket. Those feed-thru holes should be a good bit smaller. Also, unless vibration is occurring, would not plastic washers offer less stress and cause far less cracking when tighten against your rigid plastic support bracket? Next, why use metal rods if thermal conduction is an issue? Here ceramic would offer superior thermal insulation and and provide the same rigid stability. Since you are redesigning, why not consider these changes? I also should point out that thermal cycling of plastic stressed against those steel rods will create cracking; ceramic has an order of magnitude less thermal expansion over large temperature variations and just replacing those vertical steel rods with ceramic (and not over tightening those bolts) might solve part of the cracking issue in those locations.

Also, is the crystal using a thermal shield? If not, the crystal will see those RmTemp walls and create a huge thermal load. Can't see in the pic so I assume it is shielded - if not, redesign is essential ... .

Finally, might want to use your full name on the lead title post ... .

Jon Rosenstiel
Posts: 1345
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Re: High purity Germanium detector design related clarificat

Post by Jon Rosenstiel » Fri Aug 25, 2017 7:20 pm

I believe the purpose of the holes are to reduce leakage current.

Below is a pic of an Ortec item. Ortec also uses a triangular support bracket similar to the plastic one in Ashish's detector, but it appears to be made from G10 or maybe G11 fiberglass laminate material, but without the holes.
2017-08-25 11.51.39.jpg
Jon Rosenstiel

Post Reply