Chain impact mill

[snoman701]

I know it's noisy, but a steel ball mill with 3" steel balls isn't exactly quiet either. I just would like an idea if it's noisier than a ball mill if he has the two to compare noise levels.

Tough call...I'd say it's a different noise.

Looking at his design, there's some very specific things that I'd do to reduce the noise. One is increase wall thickness to 1" on the impact faces using at least AR400 steel. The chain should be the consumable, not the walls. Then I'd increase the thickness of the front and back face to at least 1/2".

Efficiency can be increased by using an octagonal shape instead of round. It will force grinding, instead of having pieces get spun around.

The design should be three bearings, not two. One inside, one outside, and one further out where it attaches to the motor. It will reduce vibrations. The front face encapsulates it, instead of having a hole. Get the whole thing sprayed with rhino lining following that.

No welding on the shaft, just get a piece of 4" diameter piece of steel, have someone drill a 1 1/2 hole for the shaft (and key it). Diametrically opposite positions on this, cut a slot for your chain. Then drill the holes for bolts to hold the chain in.

The goal is just to reduce ALL vibration except the grinding. That's where the noise comes from. Heavier it is, the quieter it will be. Steel is cheap.

Give up on the granulator?? I've wanted to buy a small one.

 

[kjavanb123]

Nice piece of equipment, what would you estimate the weight of boards that could be fed into this mill per hour?

And how noisy is it while operating?

I have not measured the throughput yet, but by just visual observation, it takes longer to mill circuit boards either depopulated or as whole, than it does quartz ore.

Had I chose a 3HP or bigger motor, it might had decreased milling time.

As for noise, it is noisy, using 12mm thick chamber would decrease the noise.

 

[nickvc]

Kevin you could wrap it in thin polystyrene sheet to deaden some of the noise but if it’s away from people the noise will not really matter.

 

[Smack]

I would say the chain/hammer mill is slightly louder. If I'm going to be in my shop when the ball mill is running, I will put a large cardboard box over it. The 5 layer gaylords really quiet it down.

 

[kjavanb123]

I would say the chain/hammer mill is slightly louder. If I'm going to be in my shop when the ball mill is running, I will put a large cardboard box over it. The 5 layer gaylords really quiet it down.

Thanks for the tip. I will try this out.

 

[kjavanb123]

All,

Here is some update on results for running and panning depopulated and whole circuit boards.

After running a kilogram of depopulated circuit boards mostly motherboards, in my hammer mill and pan copper, dissolved it in hydrochlorid acid to clean any solders, incinerate it and will melt tommorow so I can find copper recovery rate for this chain mill.

I also ran 4 chips as can be seen from picture below,


Panned the result and dissolved in hcl to remove any ferrous or tin, incinerated, follow dissolution in nitric acid, after some shaking the pan, the gold bonding wires show up,


Here is a close up,


My next experiement would be manually depopulate a kilogram of motherboards, smelt the components and record the amount of gold, silver and palladium recovered,

Then take another kilogram of similar circuit boards and run it in mill, sift and run the result on gravity separation equipment, and recover gold, silver and palladium from concentrated.

This experiement will assist in finding out milling vs smelting recovery rates.

 

[Shark]

Great job, I love seeing your equipment and how you work to make it all come together.

 

[kjavanb123]

Shark,

Thanks. I enjoy your post specially about copper electrolysis.

Today I melted down incinerated copper powder from milling and panning exactly 1009 grams of depopulated PCBs.

My furnace insulation was broken so melt did not complete, but over all 210 grams of copper was recovered.


And some left over that did not pool,


I will re melt these to get a solid copper bar to have final amount of copper recovered from a kilogram of depopulated PCBs.

But 21% copper from depopulated board seems about right.

Best regards
Kj

 

[4metals]

But 21% copper from depopulated board seems about right.

That fits right in with the 15 to 20% average copper content I have experienced. So this mill actually allowed you to skip the incineration step and go right to powder and metallics. I would bet the powder, if incinerated and fire assayed, would be holding values as well. It would be interesting to process a large enough batch and assay the fine powder.

 

[kjavanb123]

4metals,
My main goal was to skip the pyro and incineration steps, if I change my mill output from 5mm to 1mm or even 0.5mm the finer the result would be hence more copper recovery.

As far as values I highly doubt they do have any, as these were already depopulated and processed for values, what I ran for copper recovery was just bare boards with some solders on them which I cleaned by hcl.

My crucible broke upon re melting, but copper fine melted. I let them cool so I can weigh it and have an assay.

 

[4metals]

The boards you cut up to fit into the feed chute weren't totally depopulated, the good stuff was gone but there are likely values from fingers. I would think a mill like yours with larger holes would pass the metallic fraction easier and faster, then sifting is a quick and inexpensive way to get the metallics for smelting. If you had solder joints they could possibly have been soldered to a gold plated hole in the board and removing tin with HCl would take the gold stannate as well but if the solder was left in and the molten copper air sparged, you would upgrade the copper and leave just copper and PM's. Plus an acid step involves an entire environmental issue as well as handling. I'm a fire guy!

My thinking is to minimize the manual depopulating to the easy picking parts and knowing any remaining PM's can be recovered. Then a second mill with very fine mesh on the discharge will yield fine powder which can be sampled and assayed. You may be able to ship the powder once or twice but the refiner will figure out that it is still unburned circuitry and make you burn it. But from experience with powders from motherboards, they do hold values. The prospect of processing them without pyrolysis or incineration is intriguing. Without the incineration process I would think the fire assay would be very challenging to reduce.

Here in the US, the scrubbing of a smelt operation on the copper is very do-able but the burning is what the EPA doesn't really care for. The labor to manually depopulate is costly in the US but if it can be limited to easily picked components it isn't as costly if the remaining powders and copper recovered this way can be "mined" for their copper and PM values effectively. The potential of this approach is worth discussing further.

 

[IdahoMole]

Thanks Kevin for showing us the build and use of your mill, it is very interesting. 4metals got me thinking this morning, if the burning of the boards can be eliminated it would remove a huge obstacle for myself and probably many others.
Let me think out loud for a moment and see if i am understanding the theory. I could take populated boards and mill them, concentrate the material with a sluice or blue bowl etc., smelt the concentrates, pour impure copper anodes for a copper cell and process the resulting slimes for PM's?
This is all theory for me because i have no hands on experience with this particular process but the threads on smelting and copper cells have intrigued me. I did build a furnace this winter with the intention of trying my hand at smelting and now that the weather is beginning to warm up a bit i am getting excited about getting into the lab. I would like to build a copper cell soon.
I will be watching this thread and i appreciate everyone's comments.

 

[kjavanb123]

Idahomole

Thanks for your interest. I have already done a similar project with Mt Baker Mining And Metals company, where I sent them pulverized populated circuit boards and they ran it on their shaker table and assay the concentrates.

If you have not seen the thread about, please let me know so I can put it here, there is whole discussion about assay results and recovery rate.

In my opinion milling sifting then gravity separation following smelting and sparging to get 99% copper with all the PMs would work instead of pyrolysis and incineration.

The mill I built would pulverize the populated circuit board, about 90% of it pass mesh 100, and 10% over size, which can either be pyrolized or I am thinking ball milling.

Soon I will do an experiment to see how efficent this milling and separation process is, by depopulating and lead smelting 1 kg (~2lbs) of identical circuit boards, and process the same 1 kg circuit board using milling spearation and copper smelting.

I will try to post that here as I am currently shrdding my left over depopulated boards to pulverize and separate copper.

Best regards
Kj

 

[4metals]

Back in the day I incinerated boards and crushed the burnt remains and sifted off the metallic fraction. The metallic fraction was passed over with strong magnets to remove any magnetic material and the rest was copper, solder and PM's. The boards we received in had their heat sinks and batteries and, in some cases, even processors removed before we got them.

The copper fraction as we called it was smelted and oxygen sparged to slag off just about everything but copper and the PM's. Then we, at first, assayed the bars and shipped them to a copper smelter and were paid on copper and PM's. Eventually when the feed was consistent we began to run copper cells.

The powder was sifted to a -60 mesh. As a fine powder it could be sampled and shipped to a refiner in Europe who processed sweeps. The only difference I see here is the powders Kevin is producing have never been incinerated so they will not perform the same as incinerated powders and the end refiner will likely balk. Since the boards we burned had visible values on them, fingers etc., we knew some of those values would end up in the fines. But the trade off is the labor to snip off everything of value is huge, unless labor is cheap.

An option to recover the values from the powders may be to leach out the values in a Deano-ish style leach circuit. I still think there will be difficulty fire assaying the unburnt powders but since they can be sampled efficiently because of their particle size a small sample can be incinerated just to simplify the assay process.

It is an intriguing prospect, just because the incineration process is so onerous. Eliminating it entirely is, well, intriguing!

 

[snoman701]

Intriguing, yes, but is it viable on the large scale?

You've said yourself, you are a pyro guy. When you compare the efficiency of the pyrometallurgy process, on a large scale, do you really think it's reasonable to try to shred/mill/concentrate the stuff?

I can look at a lot of boards and say, "yeah, that's going to refine at $10 a lb, and I'm only getting $5", but even at $5 a lb gross profit, I'd have to refine 20,000 lbs of that board before I can pay the startup costs on a very lean professionally set up line to shred, mill, concentrate, smelt, refine. That's a huge undertaking.

I don't mind dealing in e-scrap, but at the end of the day, it still seems that the best thing to do is simply sell it upstream. I guess if I was already dealing in loads of weird, high yielding telecom, there might be a niche where the refining costs will support an operation such as this, but then I question if you could find the quantity of this product to make such an operation profitable.

But then, the above could apply to just about any business proposition.

 

[rickzeien]

Here is an interesting paper on PCB pyrolysis in molten salt.

May not be viable on a small scale but I wonder how the EPA would view this process.

https://www.sciencedirect.com/science/article/pii/S2215016115000138

Sent from my LG-H872 using Tapatalk

 

[4metals]

When you compare the efficiency of the pyrometallurgy process, on a large scale, do you really think it's reasonable to try to shred/mill/concentrate the stuff?

To even begin to start the pyrometallurgy process you need incinerate, crush and sift to get the material to melt. To use Kevin's method you would need to shred, and sift to get the material to melt.

Incineration is expensive between air permits and buying an incinerator. Entry level small incinerators go for $20K just for the unit, not talking stacks and permitting.

Chain mills (I have since looked them up after reading this thread) are available for $5K and their capacity is greater than incineration on a daily basis. http://gold-mill.com/Products.html Kevin has proven that for circuit boards they will separate the metal fraction from the boards. That is a big plus.

When I refined, the only reason I went through all of the effort to produce a sample-able powder and a sample-able copper base bullion was to be paid for what was there from the smelter / refiner. You give that up when you sell it upstream.

 

[rickzeien]

On a small scale I read about (but cannot find) a article/post for PCB pyrolysis using a Dutch kettle cast iron pot with a lid.

A pipe was threaded into the bottom center of the pot. The pipe top was a few inches from the top to allow the gas to vent into the flame from a burner that the pot sat on.

I wish I could find the link because I thought it was a genius way to simply process SMDs and PCB on a small scale.

Sent from my LG-H872 using Tapatalk

 

[rickzeien]

With the "mill and separate" via shaker table model there is still a very large fraction of the material that is hazardous waste. What is the cost and logistics involved with proper disposal of this waste stream?

My research has not yielded any suitable "buyers" for this material.

In the "burn" model the slag encapsulates the solid waste stream and is sold as aggregate.

Sent from my LG-H872 using Tapatalk

 

[kjavanb123]

With the "mill and separate" via shaker table model there is still a very large fraction of the material that is hazardous waste. What is the cost and logistics involved with proper disposal of this waste stream?

My research has not yielded any suitable "buyers" for this material.

In the "burn" model the slag encapsulates the solid waste stream and is sold as aggregate.

Sent from my LG-H872 using Tapatalk

I remember seeing a link of company in China or in East EU where they processed failed circuit boards, collected copper and fabrics turned into a brick that was used in contruction, I think Patnor101 provided the link.