Some worries about cupellation

[fireassay]

How long will take for a cupellation to get finish? Is there any way to understand that the Pb got completely removed from the dore beads.( Au)

I start to run corrected fire assay since last February. But for last three days proof samples showing high retention value. Generally, we are using two proofs for ten samples and those proofs are loaded diagonal corner each other ie 1& 12. The problem am facing right now is that, if the first proof shows the value -0.560 and the second proof shows value -0.102. Temp we are doing FA is 1050 degree Celsius and the duration is 25 minutes. I never let those cupel inside the furnace after 25 min. However i remove them and tap on the side of the cupel in order to happen the BLICK.


I could see, after removing the cupel from the muffle, dense fumes coming from the dore beads, but after tapping on the side of the cupel, blick happened and those fumes disappeared. What does those dense fumes mean? are they Pb or something. Does it mean that high retention value on the proof sample due to those suppressed dense fumes. How could i tackle this tricky situation.

We are using this algorithm to calculate proof sample;
sample taken - weight obtained

For eg A B
140.232 140.723 A = proof sample
B = proof sample after parting, cornet value.

 

[4metals]

What type of assay furnace are you using? How big is the interior? Is it a muffle or a box furnace? These things influence the time required and how many samples you can fit in at once. It also effects where the proofs should be placed.

How much does the proof in position 1 vary from the proof in position 12? What type of balance are you using? Are you weighing to 6 places? 140.232 mg?

 

[fireassay]

What type of assay furnace are you using? How big is the interior? Is it a muffle or a box furnace? These things influence the time required and how many samples you can fit in at once. It also effects where the proofs should be placed.

How much does the proof in position 1 vary from the proof in position 12? What type of balance are you using? Are you weighing to 6 places? 140.232 mg?

We are using muffle furnace with temperature range up to 1200 degree Celsius. The working temperature range is from 650 to 1150 degree Celsius. It will take 2.5 hrs to heat uniformly, and the type of muffle we are using is nitride bonded silicon carbide. Moreover, instead of cracking the door to let air at the time of oxidation, we are using a blower.The internal dimension of the furnace is 110mmx200mmx230mm (Height x Width x Depth).

Type of Temp control - K type or R type (1200°C)
senser (Cr/Al) or Pt Pt Rh
Heating element - Silicon carbide

We are using two 6 HB Mabor Cupel bullion blocks, so we could carry five samples with one duplicate each along with two proof gold samples, totally twelve.
The 1st and 12th HB reserved for proof assay samples.

We are using sartorius micro balance with sensitivity 0.001 mg, weighing as 140.232 mg

am gonna post a reading we got three day before

initial reading final reading Final - Initial
Proof G #1 139.236 mg 139.682 mg -0.446 mg
proof G# 12 139.221 mg 139.300 mg -0.079 mg

 

[4metals]

The reason I asked the furnace size is to see if you are putting too many cupels into each load but with only two 6HB cupel blocks in a furnace with a floor 20CM by 23CM you are not getting too close to the walls of the muffle.

Using the data you gave your correction factor for the first proof is .9968 and the second proof is .9994.
To correct your results, the result from each sample is multiplied by the above correction factors.

Were the proofs run for the same karat of gold? Was the identical quantity of silver added to each? Was the silver added as assay silver beads or inquarts? I am thinking the results are too far from each other. Likely the front proof, position 1, lost metal due to its position in the muffle. The rear proof position 12 is actually quite high, usually I see .9985.

Is it possible you have a hot spot in the muffle or the blower is hitting one side more than the other? Normally I like to add a row of dummy cupels to buffer the effect of opening the door and I avoid placing proofs on the ends of rows. Can you try a load with proofs at position 3 and position 10 to see if they are closer to each other? I assume position 1 is front row on either end and position 12 is back row opposite end.

Are you adding extra silver to the proof to compensate for the silver in the samples you are running? Is the 139 mg proof bead close to weight of the expected fine gold in the samples?

 

[goldsilverpro]

As 4metals did, I always kept a row of dummy cupels in the front and, sometimes, I also put them on the sides. The row in front, especially, can definitely make a difference.

The bead will be very bright and glowing and moving when lead is still in it, even with just a small amount. When finished, it will look metallic and duller but will not glow. This color change is immediate and is easily seen. The blick occurs at the point when the last trace of lead is removed. I never based cupellation on time, but always by the appearance of the bead.

 

[fireassay]

The reason I asked the furnace size is to see if you are putting too many cupels into each load but with only two 6HB cupel blocks in a furnace with a floor 20CM by 23CM you are not getting too close to the walls of the muffle.

Using the data you gave your correction factor for the first proof is .9968 and the second proof is .9994.
To correct your results, the result from each sample is multiplied by the above correction factors.

Were the proofs run for the same karat of gold? Was the identical quantity of silver added to each? Was the silver added as assay silver beads or inquarts? I am thinking the results are too far from each other. Likely the front proof, position 1, lost metal due to its position in the muffle. The rear proof position 12 is actually quite high, usually I see .9985.

Is it possible you have a hot spot in the muffle or the blower is hitting one side more than the other? Normally I like to add a row of dummy cupels to buffer the effect of opening the door and I avoid placing proofs on the ends of rows. Can you try a load with proofs at position 3 and position 10 to see if they are closer to each other? I assume position 1 is front row on either end and position 12 is back row opposite end.

Are you adding extra silver to the proof to compensate for the silver in the samples you are running? Is the 139 mg proof bead close to weight of the expected fine gold in the samples?

We are running 999.9 proof for 22K gold, and we inquartate with 2.3 times of 999 silver beads to the fine gold present in the sample. For eg, if we are running an assay for 916.6 gold, and taking an unknown sample nearly 152.326 mg, we use this algorithm to calculate how much proof sample should weigh.
152.326 times .9166 = 139.622 mg . Silver we used to add would be +/- 1 mg difference, most probably 352,353 or354 mg.

WE are adding same amount of silver ie +/- 353 to all 12 samples in each section, and i don't have any idea about the air from the blower hitting one side of the cupel or not.

I would like to add the procedure we follow along with this: Before loading, we pre burn the cupel for 15 minutes or the cupel turn to the muffle color. Then we load the sample with the charger and push the two cupel to the middle of the muffle, and close the door. We let them for 4 minutes and then let the blower for next 16 minute. Then switch off the blower in and for next 4 min we let the blower out. Normally we never let the cupel inside the furnace after the cupellation, but in order to avoid sprouting and to avoid breaking of the dore beads, time of rolling, we allow them inside the furnace, switching off, cracked the door a little bit, up to 945 degree Celsius.

Up to which value the correction value is reliable ie in between .998X to 1.001X is okay. What will happen if i get the correction value +/- in the same result for eg 0.9996 and 1.0003.

 

[fireassay]

As 4metals did, I always kept a row of dummy cupels in the front and, sometimes, I also put them on the sides. The row in front, especially, can definitely make a difference.

The bead will be very bright and glowing and moving when lead is still in it, even with just a small amount. When finished, it will look metallic and duller but will not glow. This color change is immediate and is easily seen. The blick occurs at the point when the last trace of lead is removed. I never based cupellation on time, but always by the appearance of the bead.

is the row in the front mean if it close to the door or to the other end of the muffle. Is it okay to tap at the side of the cupel, to let the blick happen. In order to save time, we used to stick with this procedure but the draw back we had noticed that while rolling the bead start to break.

 

[goldsilverpro]

As 4metals did, I always kept a row of dummy cupels in the front and, sometimes, I also put them on the sides. The row in front, especially, can definitely make a difference.

The bead will be very bright and glowing and moving when lead is still in it, even with just a small amount. When finished, it will look metallic and duller but will not glow. This color change is immediate and is easily seen. The blick occurs at the point when the last trace of lead is removed. I never based cupellation on time, but always by the appearance of the bead.

is the row in the front mean if it close to the door or to the other end of the muffle. Is it okay to tap at the side of the cupel, to let the blick happen. In order to save time, we used to stick with this procedure but the draw back we had noticed that while rolling the bead start to break.

Cracking while rolling is most likely caused by the retention of lead. Evidently, tapping the cupel before the bead is truly finished doesn't work. Maybe it blicked because it cooled enough to solidify the bead when the door was open. Also, I've never seen it done in any of the books.

I don't understand why you feel the necessity to see the "blick." The blick is just a phenomenon that often happens, not something you look for on each assay. With some conditions and/or metal combinations, there is no observable blick, or flash, as it's sometimes called. I've run at least 10,000 fire assays and have seen the blick maybe only 50 times. Each time, it was by accident. I just happened to open the door at the very instance when the blick occurred. I never looked for it. No reason to. I always went by the appearance of the bead.

Sprouting can often occur when there is more than twice as much silver as gold. Sprouting produces losses. The usual remedy is to place a very hot empty cupel on top of the cupel with the finished bead on it and let it remain in the furnace a couple more minutes. The 2 cupels are then carefully removed together from the furnace and allowed to cool before removing the top one. This works quite well.

Does the door open to the side or does it slide up and down?
Gas or electric?
Is there an exhaust hole in the back of the furnace?
How much air per minute is provided by the blower?
How much lead are you using?
How many samples do you run per 8 hr day?

 

[fireassay]

As 4metals did, I always kept a row of dummy cupels in the front and, sometimes, I also put them on the sides. The row in front, especially, can definitely make a difference.

The bead will be very bright and glowing and moving when lead is still in it, even with just a small amount. When finished, it will look metallic and duller but will not glow. This color change is immediate and is easily seen. The blick occurs at the point when the last trace of lead is removed. I never based cupellation on time, but always by the appearance of the bead.

is the row in the front mean if it close to the door or to the other end of the muffle. Is it okay to tap at the side of the cupel, to let the blick happen. In order to save time, we used to stick with this procedure but the draw back we had noticed that while rolling the bead start to break.

Cracking while rolling is most likely caused by the retention of lead. Evidently, tapping the cupel before the bead is truly finished doesn't work. Maybe it blicked because it cooled enough to solidify the bead when the door was open. Also, I've never seen it done in any of the books.

I don't understand why you feel the necessity to see the "blick." The blick is just a phenomenon that often happens, not something you look for on each assay. With some conditions and/or metal combinations, there is no observable blick, or flash, as it's sometimes called. I've run at least 10,000 fire assays and have seen the blick maybe only 50 times. Each time, it was by accident. I just happened to open the door at the very instance when the blick occurred. I never looked for it. No reason to. I always went by the appearance of the bead.

Sprouting can often occur when there is more than twice as much silver as gold. Sprouting produces losses. The usual remedy is to place a very hot empty cupel on top of the cupel with the finished bead on it and let it remain in the furnace a couple more minutes. The 2 cupels are then carefully removed together from the furnace and allowed to cool before removing the top one. This works quite well.

Does the door open to the side or does it slide up and down?
Gas or electric?
Is there an exhaust hole in the back of the furnace?
How much air per minute is provided by the blower?
How much lead are you using?
How many samples do you run per 8 hr day?

I also noticed that if i left the cupel inside the furnace for a long time after the cupellation, up to 940 degree from 1050 degree, while rolling, no cracking of the bead is observed, and i used to get a nice strip with equal side and nice edges.

We are using is an electric one, and the door is open to the side. The exhaust hole is in the back of the furnace, and attached to a scrubber.
The amount of air providing to the furnace is a tricky question, and am using the regulator of the blower to adjust it. The regulator has ten points, and normally set it in the third point. How much amount of air that needed for the oxidation process, and how could we measure the amount of air?

we are using 4 gram of lead uniformly to all twelve sample along with 353+/- 1 pure silver drops with it. Actually how much time is required for a cupellation process, and how long should i need to keep for the cooling after the end point of the cupellation? Is it necessary to open the door for the cooling phase after the cupellation and normally we switch off the furnace for the cooling purpose, to reduce the temperature.

The amount of samples we do depend on the work, but not less than 20 samples, ie 4 times per day of 7 hr work.

 

[Westerngs]

The "blick" being mentioned is the "latent heat of fusion". That is, it is the energy released as the bead center solidifies. It happens at the end of every properly conducted cupellation. We just don't see it very often because we either let the beads solidify inside the furnace or we cover the charged cupel with an empty heated cupel.

The empty row of dummy row of cupels mentioned by both GSP and 4metals is there to help maintain the muffle space behind them uniformly heated. I place them about 4 inches back from the front of the muffle and they must extend completely across the face of the muffle.

As to the gold to silver ratio, I prefer 1:4 rather than the 1:2.3 you mention using. as GSP mentions, you must be careful to keep the beads from spitting, but it can be done using either of the methods GSP mentions.

I would expect the retention to be in the 100.00% - 100.50% range and for the control sample replicates to agree within 0.05%.

Another thing you must look at is the parting procedure. What acid strength do you use and for how long? Also, do you use individual parting vessels of some type or parting racks?

please explain your parting procedure.

It seems to me that if you are doing as many samples daily as you say you are, you or someone else there should have experienced enough to figure out what is going on. Or have you just recently started doing gold fire assays? If so, who set up the initial procedure?

 

[4metals]

In a gold fire assay I have never used a positive correction factor (greater than 1) to my thinking a correction factor greater than one means you have had some gold losses because by the nature of what is happening in a cupellation you will always retain silver so the final weight must exceed the start weight. If it doesn't you have lost some gold either rolling or parting or even from spitting in the furnace.

Just out of curiosity, is the side your door opens from (opposite the hinge) the same side as where position number 1 is? If it is, and position # 1 always loses more than the position in the rear, I suspect airflow over the proof sample is excessive and effecting the result.

I also prefer 1:4 silver ratio. I find that at that ratio I can part without rolling. The gold comes out like brown coffee grinds as it breaks up in parting and one has to be careful when decanting but after annealing the gold can be easily weighed and you will save a lot of time in the procedure. I have run assays both ways, in fact the procedure involving rolling was referred to as the "long procedure" when I was taught, and the direct parting was the "short procedure." With a little practice and care you should be able to produce equal results either way. If so you will choose the short procedure because of the time it saves.

I notice you are adding equal silver weight to samples and proofs. To get the best proof possible the goal is to match the silver content in the sample to the silver content in the proof. To do this in a modern assay lab you can use an XRF result to determine the silver in the sample and add that percentage times the sample weight to the proof. This way the silver to gold ratio in the final bead is the same for the proofs and the samples. I know that Indian's seem to like to add silver to their alloys, especially 22 karat, If you do not account for that extra silver that is in your sample in your proof, your results will differ. When I run assays to settle with a major refiner, I always run a proof with each sample. I run 3 cups of sample, one dore, and a proof. (I like to put the proof in the center of each row.)

I normally use half gram samples, and I use pre made lead cups weighing 15 grams each. What does your sample weigh for a 4 gram lead charge?

Finally the "blick." I actually have seen it but it is likely because I was opening the door at the exact moment and there it was for an instant and then it was gone. It reminds me of the green flash mariners speak of in the ocean, I know that exists as well. I average 20 nights a year at sea and have been averaging that for 40 years, I often sit out to watch a sunset but I have never personally seen a green flash. Maybe they're related?

 

[fireassay]

The "blick" being mentioned is the "latent heat of fusion". That is, it is the energy released as the bead center solidifies. It happens at the end of every properly conducted cupellation. We just don't see it very often because we either let the beads solidify inside the furnace or we cover the charged cupel with an empty heated cupel.

The empty row of dummy row of cupels mentioned by both GSP and 4metals is there to help maintain the muffle space behind them uniformly heated. I place them about 4 inches back from the front of the muffle and they must extend completely across the face of the muffle.

As to the gold to silver ratio, I prefer 1:4 rather than the 1:2.3 you mention using. as GSP mentions, you must be careful to keep the beads from spitting, but it can be done using either of the methods GSP mentions.

I would expect the retention to be in the 100.00% - 100.50% range and for the control sample replicates to agree within 0.05%.

Another thing you must look at is the parting procedure. What acid strength do you use and for how long? Also, do you use individual parting vessels of some type or parting racks?

please explain your parting procedure.

It seems to me that if you are doing as many samples daily as you say you are, you or someone else there should have experienced enough to figure out what is going on. Or have you just recently started doing gold fire assays? If so, who set up the initial procedure?

Am just a rookie in this field, and am only allowed to one set of entire process in alternative days. If they find any kind of abnormalities in my result, they will repeat the test.

We are using 1.2 and 1.3 specific gravity HNO3 for the parting process ,and taking 300 ml of parting acid for each section. For the first parting, if the temperature reach 90 degree, we insert the tray in to 1.2 HNO3 and allowed to boil gently, and the time allowed in each section is 15 minutes. After 15 minutes, decant the tray and put in the second acid, in a separate beaker, when the temp reached 105 degree for the second parting and boil gently. Then wash with warm DM water and allow to dry on a hot plate for 8 min, low heat, and again anneal in a annealing furnace, 700 degree, for 5 minute.

 

[fireassay]

In a gold fire assay I have never used a positive correction factor (greater than 1) to my thinking a correction factor greater than one means you have had some gold losses because by the nature of what is happening in a cupellation you will always retain silver so the final weight must exceed the start weight. If it doesn't you have lost some gold either rolling or parting or even from spitting in the furnace.

Just out of curiosity, is the side your door opens from (opposite the hinge) the same side as where position number 1 is? If it is, and position # 1 always loses more than the position in the rear, I suspect airflow over the proof sample is excessive and effecting the result.

I also prefer 1:4 silver ratio. I find that at that ratio I can part without rolling. The gold comes out like brown coffee grinds as it breaks up in parting and one has to be careful when decanting but after annealing the gold can be easily weighed and you will save a lot of time in the procedure. I have run assays both ways, in fact the procedure involving rolling was referred to as the "long procedure" when I was taught, and the direct parting was the "short procedure." With a little practice and care you should be able to produce equal results either way. If so you will choose the short procedure because of the time it saves.

I notice you are adding equal silver weight to samples and proofs. To get the best proof possible the goal is to match the silver content in the sample to the silver content in the proof. To do this in a modern assay lab you can use an XRF result to determine the silver in the sample and add that percentage times the sample weight to the proof. This way the silver to gold ratio in the final bead is the same for the proofs and the samples. I know that Indian's seem to like to add silver to their alloys, especially 22 karat, If you do not account for that extra silver that is in your sample in your proof, your results will differ. When I run assays to settle with a major refiner, I always run a proof with each sample. I run 3 cups of sample, one dore, and a proof. (I like to put the proof in the center of each row.)

I normally use half gram samples, and I use pre made lead cups weighing 15 grams each. What does your sample weigh for a 4 gram lead charge?

Finally the "blick." I actually have seen it but it is likely because I was opening the door at the exact moment and there it was for an instant and then it was gone. It reminds me of the green flash mariners speak of in the ocean, I know that exists as well. I average 20 nights a year at sea and have been averaging that for 40 years, I often sit out to watch a sunset but I have never personally seen a green flash. Maybe they're related?

Normally, we keep #1 proof at left end side of the rear row, first depression, and the #2 proof, right end side, twelfth depression, near to the door.

We are using common parting for all samples in 4x4 quartz parting tray, using 300 ml 1.2 for first and 300 ml 1.3 for second parting. We use the help of XRF to determine the amount of gold percentage before doing fire assay, but not for determining silver content, and next time i will try it. The place i works, southwest, they add more copper to the alloy , may be 990, than silver in 22K gold, because copper is more cheaper.

 

[goldsilverpro]

Were I you, I would get some good fire assay books. Although old (both 1940), I feel these 2 are about the best. I have both in bound form. They're not expensive new - about $20 each. For a free digital copy, the 1st one, Shepard and Dietrich, can be downloaded in full by clicking PDF on the left. The 2nd one, Bugbee, my favorite, can't be downloaded here but it can be read in full on the site.
https://archive.org/details/fireassaying032695mbp
http://babel.hathitrust.org/cgi/pt?id=mdp.39015024222518;view=1up;seq=7

 

[fireassay]

Were I you, I would get some good fire assay books. Although old (both 1940), I feel these 2 are about the best. I have both in bound form. They're not expensive new - about $20 each. For a free digital copy, the 1st one, Shepard and Dietrich, can be downloaded in full by clicking PDF on the left. The 2nd one, Bugbee, my favorite, can't be downloaded here but it can be read in full on the site.
https://archive.org/details/fireassaying032695mbp
http://babel.hathitrust.org/cgi/pt?id=mdp.39015024222518;view=1up;seq=7

Thanks for your help

 

[LordGoldsThroneroom]

In a gold fire assay I have never used a positive correction factor (greater than 1) to my thinking a correction factor greater than one means you have had some gold losses because by the nature of what is happening in a cupellation you will always retain silver so the final weight must exceed the start weight. If it doesn't you have lost some gold either rolling or parting or even from spitting in the furnace.

Just out of curiosity, is the side your door opens from (opposite the hinge) the same side as where position number 1 is? If it is, and position # 1 always loses more than the position in the rear, I suspect airflow over the proof sample is excessive and effecting the result.

I also prefer 1:4 silver ratio. I find that at that ratio I can part without rolling. The gold comes out like brown coffee grinds as it breaks up in parting and one has to be careful when decanting but after annealing the gold can be easily weighed and you will save a lot of time in the procedure. I have run assays both ways, in fact the procedure involving rolling was referred to as the "long procedure" when I was taught, and the direct parting was the "short procedure." With a little practice and care you should be able to produce equal results either way. If so you will choose the short procedure because of the time it saves.

I notice you are adding equal silver weight to samples and proofs. To get the best proof possible the goal is to match the silver content in the sample to the silver content in the proof. To do this in a modern assay lab you can use an XRF result to determine the silver in the sample and add that percentage times the sample weight to the proof. This way the silver to gold ratio in the final bead is the same for the proofs and the samples. I know that Indian's seem to like to add silver to their alloys, especially 22 karat, If you do not account for that extra silver that is in your sample in your proof, your results will differ. When I run assays to settle with a major refiner, I always run a proof with each sample. I run 3 cups of sample, one dore, and a proof. (I like to put the proof in the center of each row.)

I normally use half gram samples, and I use pre made lead cups weighing 15 grams each. What does your sample weigh for a 4 gram lead charge?

Finally the "blick." I actually have seen it but it is likely because I was opening the door at the exact moment and there it was for an instant and then it was gone. It reminds me of the green flash mariners speak of in the ocean, I know that exists as well. I average 20 nights a year at sea and have been averaging that for 40 years, I often sit out to watch a sunset but I have never personally seen a green flash. Maybe they're related?

Hey partner! Not an assay expert at all, but you got me intrigued with the "Green Flash". I have captured this with a camera ONCE and was ever so lucky to do so. I've seen it maybe 3-4 times in my 40+ years of life.... and it is quick and fleeting and will leave you going "was that it? yeah, I guess that was it!"

Essentially it is an atmospheric phenom that occurs as the VERY LAST sliver of sun sets on a flat horizon. It occurs for all of a second or two but it is pretty awesome when you see it. The "green flash" is dependent on a variation in astronomical refraction near the horizon, AND it is WAY better when there's an atmospheric temperature inversion to allow the light to "prism" through the atmosphere in the form of a mirage just above the setting sun (i.e. sea is warmer than the air above the horizon). Its remarkable when it happens but like most great things in life, its a brief and fleeting climax. (Not to mention staring at the sun is bad). At best I would say you are going to see the green for 1-3 seconds (not too different from your cupellation blick I guess?)... come to think of it, every time I have seen the green, it was summertime, the night was warm and humid but with a slight cool breeze, and I could watch the sun set over a flat horizon (the sea or over a bay). I hope you get to see it sometime soon! You'll know it when you see it,...oh, and a zoom lens of a digital camera recording can be used to "blow it up" to larger size and rewatch it if you do happen to capture it. I'd say record the setting sun every night for a week when you are at sea, you should be able to catch it. Great examples on Youtube as well.. --- and---

Ok, sorry to interrupt your discussion on cupellation.