Sciencemadness Discussion Board

Impact Sensitivity Apparatus?

UndermineBriarEverglade - 27-9-2024 at 07:13

I have built a hammer drop apparatus to test the sensitivity of ETN mixes, but am getting strange results. The apparatus consists of:



To test, I place 50mg of ETN on the "anvil", resting the striker on top of it, then draw back the hammer a certain distance. It drops down and crushes the sample between the striker and anvil. The whole thing sits on a plank of wood on my wooden floor.

But over 10 tests, I have only had two detonations: at 150 and 180cm (the max height). If the apparatus is working correctly that'd be 23J, which is way more than the reported 50% impact sensitivity of ETN (something like 5J). So something is going wrong. I greased the iron pipe to reduce friction. Perhaps my striker and anvil are not flat or parallel enough, or the contact area is too large? Maybe I need a big steel plate for the anvil to sit on? I'd appreciate any ideas, especially from people who have set up similar test equipment.

Nemo_Tenetur - 27-9-2024 at 22:31

This is my experience with my "ghetto style" IS measurement, posted lest year in exotic primaries - complex salts:

"Yesterday I´ve tried to set up an improvised "BAM-Fallhammerapparat" to measure the impact sensitivity of my sample, but after several trials I gave it up. This is nothing you can do with "ghetto style" equipment and expect reliable results. An internet search revealed that it is really a challenge to get an exact and reproducible value. Even the BAM federal agency in Germany (with state of the art equipment) emphasize the trouble and problems in this area ("Herausforderung für die Qualitätssicherung"):

https://opus4.kobv.de/opus4-bam/frontdoor/index/index/docId/...

UndermineBriarEverglade - 27-9-2024 at 23:19

Thanks Nemo. I actually saw your post but couldn't figure out where to get access to that document. Do you have the PDF?

I tried sandwiching the sample between 120 grit sandpaper with no apparent increase in sensitivity. Also tried putting some bricks below the apparatus with no improvement. If it is a problem with the loose-fitting striker not being exactly parallel to the anvil, or the faces not being flat enough, I guess I could get access to a lathe and make a striker that slides into the anvil like the attached image.

anvil and striker.png - 8kB

Nemo_Tenetur - 28-9-2024 at 01:42

The complete document is restricted, available only within BAM federal agency network in Germany. Please read also the post from microtek and this oblique pendulum test with sand paper, like a hybrid between impact and friction testing.

And please don´t forget the other dangers associated with electrostatic discharge, breaking of large crystals, insufficient purity and stability etc.

Axt - 29-9-2024 at 11:30

I suspect it's the anvil having too much give, it must be rock solid and very secure. Mine had a 1"x15" steel base plate and would reliably fire PETN at a 45cm drop every time. I only used 5cm increments, trying to go get a more precise measurement would give you a headache. I wouldn't bother with trying to calculate joules, just use it as a comparative measure specific to that apparatus.

UndermineBriarEverglade - 1-11-2024 at 18:57

I found that my "anvil" had deformed after several tests, so I removed it and put the sample directly on a small steel plate. I was able to detonate ETN by whacking the striker with a sledgehammer instead of the pipe, so I figured the problem was insufficient weight. I filled the pipe with lead to a weight of 3kg, but even from a height of 150cm I still can't get reliable detonation. Not sure what to do now.

Axt, can you describe the rest of your apparatus? I'd rather not buy a steel slab but if that's the only meaningful difference...

[Edited on 2024-11-2 by UndermineBriarEverglade]

Axt - 1-11-2024 at 20:12

Video attached, your description sounds the same. This rig was scrapped probably 10 years ago now, if you were to picture the ideal one in your head it wouldn't look like this but was made with what was lying about. It was consistent though; I want to say it was a 3kg weight but cannot remember for sure. Another possibility is your tube is too tight of a tolerance, if it's too tight or flexes too much it will be slowed by the vibration of its release.

First vid is PETN firing at 40cm-45cm second is MEKP firing at 0-5cm. Interestingly AN-MEKP wouldn't fire at all up to 1m yet would fire when hit with the lowest calibre rifle, showing that bullet impact is far more influenced by friction sensitivity than impact sensitivity. There's a surprisingly weak correlation between impact and friction.



[Edited on 2-11-2024 by Axt]

Attachment: droptest_1408x1152.mp4 (6.7MB)
This file has been downloaded 293 times

Axt - 1-11-2024 at 20:18


Second

Attachment: mekp_2_compress.mp4 (4.8MB)
This file has been downloaded 264 times

UndermineBriarEverglade - 7-11-2024 at 09:56

Thanks Axt. Nice videos. The biggest difference is your apparatus isn't using a separate striker and it looks like you were testing larger samples. I'll try removing mine. If that doesn't work I'll measure the speed of the pipe to make sure PVC flex isn't impeding it. Otherwise, time for a big steel slab.

Axt - 6-12-2024 at 02:24


So you are placing the explosive of top of the anvil then a weight on top of that? If so that is the problem, think of the inertia that has and how much it will take to get up to speed, it will have a dramatic effect on the sensitivity. If you are using a "firing pin" of sorts, you'll need to keep it as light, rigid and as lower surface area as possible.

I was thinking how you'd make a dual purpose impact/friction tester, such as lever to convert the vertical to an oblique strike that can be folded up out of the way like attached.

Screenshot 2024-11-08 171607.png - 102kB

[Edited on 6-12-2024 by Axt]

UndermineBriarEverglade - 6-12-2024 at 09:57

I removed the "striker", cut out a circle of steel plate, and attached it to the front of the pipe. Surprisingly it didn't help. I then recorded a slow-motion video of the impact. The "hammer" pipe is reaching the expected speed, so friction in the pipe isn't a problem. But the hammer bounces several cm off the steel plate, and the whole setup - plate, base, and PVC pipe - jumps a smaller distance with it. I think I need a higher mass anvil, equal or greater to that of the hammer.

That lever is a neat idea.

markx - 1-1-2025 at 11:33

Just use a free falling weight dropped by a release mechanism avoiding any guiding tubes or other restraints during the decent. And a very heavy stable bottom anvil, this way you shall not encounter random deviations caused by friction and ensure consistent transfer of energy to the sample. A basket around the sample holder shall catch the weight (a hammerhead, block of metal or any other weight of choice) so it does not bounce off the apparatus and fall on your toes upon completion of the test ;) I've had quite consistent results with such a device.....granted these results shall not be comparable 1:1 against a standardized measuring apparatus, but you will be able to establish trendlines and compare relative sensitivities between samples.

Microtek - 2-1-2025 at 08:42

But how do you ensure a square impact? If you were to use a hammerhead, it could easily spin just a tiny fraction of a degree, which would be enough to make it impact the anvil but not the sample, at least fully.

markx - 2-1-2025 at 11:09

Quote: Originally posted by Microtek  
But how do you ensure a square impact? If you were to use a hammerhead, it could easily spin just a tiny fraction of a degree, which would be enough to make it impact the anvil but not the sample, at least fully.


I use such a sample holder in the anvil. It protrudes from the bottom and makes for an easy target. Works very reliably...



DSCF0614.JPG - 2.1MB DSCF0615.JPG - 2MB

Microtek - 2-1-2025 at 14:29

Ah, I see. I thought you meant to let the dropping weight fall directly onto the sample. I assume the downward face of the protruding part is machined to be precisely square and very flat.
I have seen other labs use a falling steel ball that is imparted some spin. This gives a hybrid friction/impact test.

markx - 3-1-2025 at 10:21

Quote: Originally posted by Microtek  
I assume the downward face of the protruding part is machined to be precisely square and very flat.
I have seen other labs use a falling steel ball that is imparted some spin. This gives a hybrid friction/impact test.


The faces were cut in a lathe, but without any super precision or flatness....it receives a fair beating by the hammer, so small deformations are in the equation. If I remember correctly then the sample holder was made from parts of a moped's transmission. The striker being a piece of a transmission shaft and the round part was a gear shifting element that was seated on said shaft. Conveniently the gear shifter also had radially placed channels in it to vent gases away from the sample well.
Although this setup is simple and reasonably functional, it also is big heavy and very loud even without any samples going off. I actually had plans to devise something more compact and elegant.... adjustable spring tension driven e.g. But alas, life threw other obstacles in my path and a plan it has remained so far. One day perhaps...:D

Axt - 29-1-2025 at 03:05

I just realised there is an OTC impact apparatus sitting under our nose the whole time, the 1 tonne detonator presses that are cheaply available on all major online retailer's double as a very good drop test suitable over the primary explosive range, but won't fire the likes of PETN at least not without adding weight. It has 14 teeth available spaced 8mm apart covering a range for example of cyanuric triazide at 3 teeth drop to silver nitrotetrazole at 8 teeth drop, it allows for standardised amateur testing.

https://www.youtube.com/shorts/voBkB1DrXL0

pdb - 29-1-2025 at 03:44

Bravo on the find! It reminds me of Allan Poe's letter. We're counting on you for sensitivity measurements of other primaries. I suppose you conducted multiple trials on each one to calculate an average?

Axt - 29-1-2025 at 05:57

I'm not going to lie and say I performed some amazing statistical analysis on those numbers, for example "3" means it would mostly fire at 3 and sometimes at 2. I only just threw them under it in the couple hour prior to posting here.

In the video it is TATP failing at 1 tooth and 2 tooth then firing at 3 tooth. 3 means the 4th tooth down as the first tooth is at 0mm thus it's falling 3 teeth. Then its large crystal NAP failing at 6 and firing at 7.

The CTA was also large crystal as attached.


[Edited on 29-1-2025 by Axt]

cyanuric triazide crystals.jpg - 537kB

Microtek - 29-1-2025 at 09:27

That is a nice idea. I have been meaning to get myself one of those for pressing, though I'm also considering a hydraulic variant which would make it easier to measure how much I'm pressing things. It would obviously not work in the suggested capacity.

Axt - 29-1-2025 at 22:22

Hydraulics will definitely offer more control, you could integrate a load cell into the det case holder for use with the arbour press but since it doesn't have load holding it'd be hard to be consistent with the pressure. I haven't seen a bench top one with gauge for less than a few thousand though. Other than that, you are looking at a big frame mount.

I put DPPE-1 under the 1 tonne press, it fired at 4 teeth. (1st test fired at 4, then 5 fails at 3, then fired again at 4.

You could get the 3 tonne arbour press from same product line to test secondaries, that thing is much bigger then the 1 tonne.

[Edited on 30-1-2025 by Axt]

Axt - 29-1-2025 at 23:39

This is where we are at.

drop test teeth.jpg - 1.3MB

pdb - 30-1-2025 at 01:24

If it's not too much to ask, could you test AgN₃? And DPNA if you still have some, which is pretty sensitive ? And Ag-azotetrazole ? And... and... :P

Axt - 30-1-2025 at 03:36

Dinitrobenzenediazonium perchlorate failed at 3,4,5 then fired at 6. Truly I'm not just trying to fill in every free tooth it just the way it's happening. I'll get a more conclusive number later retesting it at 5, I'm making too much noise atm being near midnight. This stuff fired at a really gentle hammer blow before, so I was expecting it to be higher. I'll have to make more of the mononitro variant.

Yeh I'll do lead azide as a standard to compare the others to, but don't have any on hand although that's a simple fix. I've never made Ag-azotetrazolate, that'd be permanganate oxidation of 5-ATZ correct?

[Edited on 30-1-2025 by Axt]

pdb - 30-1-2025 at 12:07

Yes, typically: 10 g of 5-ATZ monohydrate in 50 mL of 15% NaOH. Heat to 65°C, then slowly add 10 g of KMnO₄, keeping the temperature in the 70–80°C range. Add a few ml of ethanol to neutralize any excess KMnO₄ ( color should turn yellow). Continue stirring while raising the temperature to 90°C for 10 minutes, then vacuum filter the mixture. Rinse the precipitate with hot water and place the filtrate in the fridge for 24 hours. Finally, collect the gold-like crystals.

The Ag salt is easily obtained from aminotetrazole (AT) and AgNO₃—IMO it is extremely powerful (I have to compare to DPNA).

Axt - 1-6-2025 at 12:16

Lead azide produced one firing at max of 14.

Silver azotetrazolate was 3

Copper 4-aminotriazole perchlorate was 10

Nickel aminoguanidine bromate was 6

1,4-Diiodylbenzene was 13

Hexamethylenediperoxidediazine was 14.

Iron and gallium DMSO Perchlorate were 8

Both lead azide and Hexamethylenediperoxidediazine would be higher, they infrequently would detonate at the maximum drop height. Iron DMSO Perchlorate was sensitive though weak, the immediate part struck would explode but the explosion wouldn't propagate through the rest resulting in a feeble puff, gallium salt did the same.


[Edited on 1-6-2025 by Axt]