Jump to content
Klaas remmen

Georgian 'Bulat' technology by Zaqro Nonikashvili

Recommended Posts

As promised new pics from a new sample of my large bar of bulat. This is the same steel as the macrophoto's in post #91, and the same as the pics I posted with the graphite.

This sample was taken from the other end of the bar, which never got 'too' hot (always forged at orange)

[

If compared to the pics I made of samples in earlier stage, I conclude that the following has happened;

- The blocky cementite is broken up more and getting more spheriodised

- cementite is more closely packed

- The banding is more linear

- The space between the bands is smaller

 

any thoughts are welcome!

 

Klaas,

I agree with your description of what is happening ( has happened ). If this were my steel I would wonder...did the large cementite fragments ( the ones which are now getting broken into bits ) form during solidification, or as a result of "diffusion".....what can I do to get to those earlier in the process by heat treat or by forging.

 

Jan

Share this post


Link to post
Share on other sites

Jan, I have made metallographic pics of a few ingots. If I compare these with pics I have seen in literature (cover of David Scott's 'Metallography and Microstructure of Ancient and Historic Metals', M L Wayman, G.J., Crucible steelmaking in Sri Lanka. Historical Metallurgy, 1999. 33(1): p. 26-42.) It always consists of pearlite, needle cementite and blocky cementite. larger magnification shows dendrites and in my own examples I see that the blocky cementite is in these interdendritic regions.

I have had ingots with pearlitic matrix with veryvery small graphite in interdendritic regions in stead of cementite. I do not have had time to forge these, but who knows that works too and forms cementite upon roasting and forging...

Edited by Klaas remmen

Share this post


Link to post
Share on other sites

Jan, I have made metallographic pics of a few ingots. If I compare these with pics I have seen in literature (cover of David Scott's 'Metallography and Microstructure of Ancient and Historic Metals', M L Wayman, G.J., Crucible steelmaking in Sri Lanka. Historical Metallurgy, 1999. 33(1): p. 26-42.) It always consists of pearlite, needle cementite and blocky cementite. larger magnification shows dendrites and in my own examples I see that the blocky cementite is in these interdendritic regions.

I have had ingots with pearlitic matrix with veryvery small graphite in interdendritic regions in stead of cementite. I do not have had time to forge these, but who knows that works too and forms cementite upon roasting and forging...

 

 

Thanks Klaas,

I took the cover off an old low powered microscope this afternoon and I bought a colposcope at a garage sale a few years ago (20x). The colposcope has an analog Panasonic camera and I have to buy an analog to digital converter (mac) to capture the pictures. In both cases I may just end up taking a picture through an eyepiece. We are still looking at low power in both cases. Maybe we can compare (low power notes). I have prepared two small and two large crucibles for some remelt.

 

By the way would you send me the name(s) of some museums in your area which may have some wootz on display..I may be heading to that area one day ( unfortunately I will mis the Georgian event ).

 

Jan

Share this post


Link to post
Share on other sites

Klaas Hi.

 

Image is planar right?

How about transverse " spine "

 

I do think that it looks just like you said.

And it looks better in terms of ( more Cm spheroids in matrix )

 

Niko

Share this post


Link to post
Share on other sites

Klaas Hi.

 

Image is planar right?

How about transverse " spine "

 

I do think that it looks just like you said.

And it looks better in terms of ( more Cm spheroids in matrix )

 

Niko

 

No this pîc is transverse. I am going to make a planar sample as well, because that should have bigger spacing in banding due to stretching and not compacting (I think).

 

Jan, in this area I do not have knowledge of any blades on display. But if you are in Europe and see the chance to go to London be sure to visit the Walace Collection, the British Museum and Victoria and Albert. In Solingen in Germany (230km from Antwerp, where I live) is a well known blades museum. In Antwerp they just opened a hughe new museum with a lot of beautiful weapons, but I have not visited it yet and don't know about Wootz in there. In Brussels you have the Royal Museum of Arms with exactly 1 beautiful Russian (Kozak) Bulat saber on display.

 

If you are planning to come this way please contact me, maybe we could meet up. That counts for everyone here. :D

 

Thank you all!

Edited by Klaas remmen

Share this post


Link to post
Share on other sites

Hi Jan

 

argh, thats too familiar ! can't say as to why it does that

- it may just be that its feeling grumpy that day and will teach us another humbling lesson

- but i do bandsaw off the cracky parts and have a go at the center ...and sometimes you'd be surprised that it'll hold up and make a nice little blade.. from a spark test, does it look high? like around 2%

 

the cast iron is just unreal..what a strange and interesting looking bloom... like bloom art !!

 

 

 

Hi Klaas, very good looking steel now... surprised at the difference between that and the other piece with graphite... (thought i read somewhere that carbide can turn to graphite under certain conditions ...i think??)

- the spheres do look like some of the micro pics i've seen on some of the studies.. nice !

 

as for the big blocky carbides... once they are there.. i would think you need a heat treatment and hold time above Acm to dissolve them... then many cycles bellow Acm to reform the small spheroids ..... as i understand it, the large particles take longer to dissolve on heat cycles, and due to their larger surface area ...they will grow at a quicker rate ...... so i don't believe the low temp cycles (bellow Acm ) will remove them or shrink them in preference of small spheres . ( or maybe i'm wrong ??)

 

 

GReg

Share this post


Link to post
Share on other sites

 

 

 

Klaas,

Thanks for all the recommendations and the invitation..as soon as things are scheduled I will let you know. I will bring along some samples for your microscope ( if you see any along the way you want for your reference samples, let me know and I will cut and save) . Regarding the outside surface of the donut tek sample you made...you indicated the outside was much more worked than the hole surface..are any of the pictures you are showing of the outside of that piece?

Jan

 

 

Greg,,

My iron making is going about as well as my wootz. I placed some rice hulls at the bottom of my furnace to keep the "bloom" a little warmer...and I pointed the tuyere slightly down from the normally horizontal. Soon I discovered the furnace was sort of plugged.

I just tried to finish knowing the situation...When I could blow no more, I poked into it to see how high my bloom was sitting....nothing...panic...bad language etc.. I just left it for cleanup the next morning. Based on the next morning's find, I must have had a puddle of cast on the remaining ash and I was poking holes through it ( the sample in the photo) . Back to basics for me.

The dark metal layer is decarb and it did remove faster from the bottom of that sample..so the unincorporated iron possibility is not there. I do not know what i am putting into each crucible, the cast iron I make is as variable as the bloomery iron some is hypoeutectic by a lot (usually white) , some is eutectic and gray. I need to build up a large inventory and crush it and mix it...then I will have a relatively constant C content. Before I do that I have to give the process described here a try and it will be my next test before remelting..wish me luck.

The plan is to run at about 2600 C...iron+ carbon, no silica.

 

For those reading unfamiliar with cast iron, eutectic cast iron is about 4.5% C hypoeutectic cast iron runs from abour 2%c to 4.5% C .

Jan

Share this post


Link to post
Share on other sites

Hi Greg

 

I believe these blocky parts can break up more with more cycles at a higher temp. I will try to focus on that for this piece. The thing that strikes me is that in historical pieces the bands are way more closely packed. the graphite can indeed decompose out of cementite.

 

Jan, this donut was different. The sides of the bar are the top and bottom of the ingot. So not so much deformationdifference. The samples were took from the outsides of the bar, so at same scale of deformation.

 

 

I got news from Gotscha. He contacted dr. Badri Amaglobeli, a Georgian scientist. This man wrote a PhD on Bulat in 1984. This was not accesable due to USSR issues. Mr Amaglobeli kindly approved to share this PhD with others who are interested. Check it out on Gotscha's site: My link

It has a lot of metallographical analysis of high quality Persian blades.

Share this post


Link to post
Share on other sites

the original link seems to have been taken away,

I hope many of you where able to get to the theses before that happened.

It looked like to me to be a good deal of information on bulat/wootz

Share this post


Link to post
Share on other sites

the original link seems to have been taken away,

I hope many of you where able to get to the theses before that happened.

It looked like to me to be a good deal of information on bulat/wootz

 

Yes I don't know what happened. I got a message that the article is only available via the webpage you get to see now or by contacting Gotscha.

Sorry for the misunderstanding

Share this post


Link to post
Share on other sites

Hi

 

I hardened the blade in 810°C (oil), and tempered in 240°C

I etched in the ferric chloride, the short etching and not wash settlings gave the pretty effect.

But after washing off settlings, not well. One melting plans yet this method.

 

dsc00066h.jpg

 

dsc00067i.jpg

 

---

Adam

Share this post


Link to post
Share on other sites

Hi

One melting plans yet this method.

---

Adam

 

Adam,

Thanks for the details and the pictures...still looks good to me. Do you have a plan to avoid the unincorporated metal in your next melt? I am planning my (Georgian) melt as well, and that potential problem is something I am thinking about. It may take a few melts ( around 800) to hone in on the target Carbon concentration.

 

Jan

Share this post


Link to post
Share on other sites

Jan

 

I am planning to use normal welding electrodes, with graphite and longer melting time, about 2-2,5 hours.

And one melting 3,5 hours.

 

---

Adam

Share this post


Link to post
Share on other sites

Hi Adam

 

Nice looking blade. Tell me, why do you want longer melting times? what is the theory behind that? I have heard about Anosov and his long meltings, but I don't get why. If the metal is molten it is molten, right?

 

here is a close-up of Zaqro's steel for Gotscha's project I showed you before. The steel was made rapidly, and does not have a very 'spectacular' pattern EDIT: This in comparison to his other work

bulatZaqro.jpg

Edited by Klaas remmen

Share this post


Link to post
Share on other sites

Hi Adam

 

Nice looking blade. Tell me, why do you want longer melting times? what is the theory behind that? I have heard about Anosov and his long meltings, but I don't get why. If the metal is molten it is molten, right?

 

here is a close-up of Zaqro's steel for Gotscha's project I showed you before. The steel was made rapidly, and does not have a very interesting pattern.

bulatZaqro.jpg

 

Klaas,

Thanks for posting the detail on the blade, I love what looks like enamel, and the steel looks really good to me. I do not mean to respond for Adam but here is my thought on the theory ( using iron carbon phase diagram ).

 

If one is carburizing his/her steel at a given fixed temperature , let's say 2550 F or about 1450 C the liquid forming will be a "fixed" % carbon ( in real life I do not believe that to be always the case ) let's say about 1.5% C . If this is mixing with low carbon iron at that fixed temperature..the low carbon iron will not mix/melt. So, not only does one have to run for a long time, he/she has to bring the temperature up at the very end to get a good melt.

There are other pathways to that desired result but that is one (my current plan).

Jan

Share this post


Link to post
Share on other sites

Klaas wrote: Nice looking blade. Tell me, why do you want longer melting times? what is the theory behind that? I have heard about Anosov and his long meltings, but I don't get why. If the metal is molten it is molten, right?

 

Well, one thing is: the longer the Iron is very hot or in liquid phase it is able to take up more carbon! One idea is also that by using longer melting times, the amount of carbon in the smelt does "mingle" better unifomely into the smelt and that might have a positive influence on the inner structure of the bulat/wootz.

Think of it as in percentage of Fe3-C's in the charge, the more ironcarbides in the melt, the more of them are forming along the Va-carbides, or manganese or niob or whatever you use

it is good to look at the iron-carbon phase, where you can see the percentage of Ferro-carbides increase with the amount of C in the melt - it will be at around 30volume-% when there is 2% of Carbon in the steel and only 20 volume% when there is 1,3%C.

and you will need less energy to keep the steel in the melting phase, since the amount of C in it decreases the line where the smelt starts

there are a couple of interesting points right there

and then you have the glas/slag on top, reacting also taking up "impurities" from the smelt...

 

hmm.. there might be more about it :wacko:

Edited by Jokke

Share this post


Link to post
Share on other sites

Hi Jan, that is enamel. One of the most pure & precious of Georgia.

Original it came from Brisante, and was brought to Georgia in the 9th century. It is mineral-based and the highest quality (like this here) has a melting point at about the same temperatures of silver. That's why it is so difficult to make. It is also known as French Limuge Enamel.

 

Jan& Jokke; about the longer melting, seems to make sense... But I don't know, I had a really big piece of Armco in a crucible with cast iron, and got it out after... 20min or so. it seemed completely molten, but when I grind it in two parts, there was a piece of pure iron still inside. I posted a picture of it somewhere here. The piece of pure iron was about half the size of the original Armco piece. If this was 'dissolved' after only 20min in molten cast iron, then why wait 2,5hours to get it 'uniform'? Just a thought...

 

I got a new photo, take a look at that magnificent etch....

Bulatcloseup.jpg

Edited by Klaas remmen

Share this post


Link to post
Share on other sites

 

I had a really big piece of Armco in a crucible with cast iron, and got it out after... 20min or so. it seemed completely molten, but when I grind it in two parts, there was a piece of pure iron still inside. I posted a picture of it somewhere here. The piece of pure iron was about half the size of the original Armco piece. If this was 'dissolved' after only 20min in molten cast iron, then why wait 2,5hours to get it 'uniform'? Just a thought...

 

Klaas,

In your case the total amount of Armco iron would probably have mixed in very quickly, but only if the temperature was at or above the melting point of the final composition. Regarding long melt times, porous melts are so discouraging and wasteful, that another 45 minutes in the furnace may actually be economical....once a routine is developed one can back off if possible. I will be using organic matter in place of silica in my crucible and sometimes this gets trapped and insulation is formed inside or under my melt..that in combination with an insulating crucible makes one want to take a longer time.

 

Jokke,

I am not completely clear on your explanation but I will attempt to look it up.

Jan

Edited by Jan Ysselstein

Share this post


Link to post
Share on other sites

As promised new pics from a new sample of my large bar of bulat. This is the same steel as the macrophoto's in post #91, and the same as the pics I posted with the graphite.

This sample was taken from the other end of the bar, which never got 'too' hot (always forged at orange)

monster3goed5X.jpg

monster3goed10X.jpg

monster3goed50X.jpg

 

If compared to the pics I made of samples in earlier stage, I conclude that the following has happened;

- The blocky cementite is broken up more and getting more spheriodised

- cementite is more closely packed

- The banding is more linear

- The space between the bands is smaller

 

any thoughts are welcome!

 

Klaas,

I have the little USB microscope ( 20x-200x, older model ) and I thought I would post a picture of a wootz bar which needs one more thickness reduction to 50% of what it is now. The scale compares to the second photo in this post. Once forged down , I will develop pattern and show it again.

.....I have chaged the color tone of the photo. This bar was played with, trying for the DET transformation.

Jan

 

Edited July 26,2011

 

Klaas,

Here is a photo of the bar posted above, on this post, but forged only in the longitudinal direction. The width of the photograph is 7mm (7000 microns). I am happy with the overall look but lost the material giving me a nice black etch...so I will subject it to some high risk behavior and attempt to get it to etch black.I also wish I had forged more width on the bar.

In stretching this bar I found out I do not need 1500 grams per sword blade but 1000 grams may do and 1200 grams is more than enough.

 

 

Edited by Jan Ysselstein

Share this post


Link to post
Share on other sites

Looks similar to that second photo indeed. What side of the steel is this? is this a picture from te cross section or the 'skin'?

Share this post


Link to post
Share on other sites

Looks similar to that second photo indeed. What side of the steel is this? is this a picture from te cross section or the 'skin'?

 

Klaas,

It is the surface not the edge...I currently have no way of measuring and estimate by dividing the overall width of what is in view by the number of elements in the view ( I do this with my pocket microscope). For this area I was looking at 1/4" or 6mm and saw 20 units and guess they are about 300 micrometers wide on the average.

I am playing with the USB microscope but its limitations are serious...I will be able to document the appearance of structures for future ref..

Jan

Edited by Jan Ysselstein

Share this post


Link to post
Share on other sites

Jan

 

I am planning to use normal welding electrodes, with graphite and longer melting time, about 2-2,5 hours.

And one melting 3,5 hours.

 

---

Adam

 

Adam,

A crucible is sealed and ready for firing, it contains 15 pieces of cold rolled steel weighing about 80 grams each ( no micronutrients) . In addition to the iron, the crucible contains Ferro frit 3110 ( melting range 920-950 C ) and charcoal. The trial scares me as I have had many bad experiences when glass was added to my crucibles....the porosity scares me as well. I imagine the pieces standing up to their knees in molten glass while the top two thirds is picking up carbon and beginning to flow in to the molten glass( and my crucible is dissolving) . As long as the results continue to make sense I will keep trying some tests ( many possible advantages here) along this Georgian methodology. If I am struggling I will make a few attempts skipping the glass and trying to control carbon content by other means. If the results of this test are photogenic, I will post them....fuel is gas. Since I have no way of reaching into this crucible it will heat for two hours at heat ( 2.5 hrs in the furnace) and quick cool ( just looking for melting and carburization characteristics) .

 

Jan

Share this post


Link to post
Share on other sites

Adam,

A crucible is sealed and ready for firing, it contains 15 pieces of cold rolled steel weighing about 80 grams each ( no micronutrients) . In addition to the iron, the crucible contains Ferro frit 3110 ( melting range 920-950 C ) and charcoal. The trial scares me as I have had many bad experiences when glass was added to my crucibles....the porosity scares me as well. I imagine the pieces standing up to their knees in molten glass while the top two thirds is picking up carbon and beginning to flow in to the molten glass( and my crucible is dissolving) . As long as the results continue to make sense I will keep trying some tests ( many possible advantages here) along this Georgian methodology. If I am struggling I will make a few attempts skipping the glass and trying to control carbon content by other means. If the results of this test are photogenic, I will post them....fuel is gas. Since I have no way of reaching into this crucible it will heat for two hours at heat ( 2.5 hrs in the furnace) and quick cool ( just looking for melting and carburization characteristics) .

 

Jan

 

 

Adam,

The test went as planned and the crucible has survived...cooling as I write ( and I got a free haircut and eyebrow trim ) . I have no idea what to expect...however, it will not take much to make a convert out of me. As I proceed with this process the advantages are becoming clearer. In pevious experiments I approximated the take-up of 200mg of carbon/ per square inch/per hour at about 2000 C. Having 15 strips of 1/8"X1"X5.5" ( each having about 9 square inches of surface exposed above the glass ) I am assuming about 100 Square inches of available surface area ( these strips were tightly set against one another..reducing exposure). So 100 inches of effective surface should absorb about 20. grams per hour of carbon. Twenty grams of carbon should give a 1.6% C iron ..( based on guess work and estimates ). Now it is off to checking if glass melted, glaze sealed and flowed and if the carbon cooperated. Every time I have posted without first seeing the results it has been a disaster. If for some reason the trial is successful it will be because with a closed crucible I was able to take the operator out of the process ..all I needed was heat, I worked in a nearby location listening to the sound of the furnace.

 

By the way I used 2kgs of propane +/_ .5 kg. Here is a picture of what I put into the furnace( glass frit is a white powder and the carbon a black powder, these are not shown ) . The glaze on the strip is an estimate of glass level at the point when most of the glass has melted.

 

 

Edited by Jan Ysselstein

Share this post


Link to post
Share on other sites

Jan

 

This seems very well to be planned. I hope that you will take out from the crucible beautiful cake.

I made two melting in last week. One unsuccessful, holes inside cake. Furnace not worked well, too low temperature.

Second melting seems good, both to run above two hours. I hope that forging will run well.

But this in the next week, I have to finish Giermundbu helmet to do.

 

---

Adam

Share this post


Link to post
Share on other sites

Jan

I hope that you will take out from the crucible beautiful cake.

---

Adam

 

Adam,

Thank you and much gratitude to Zaqro Nonikashvili , Klaas and Seerp for placing this method on the forum....I am a convert. I have been making cast iron from ore, and will continue to do so for a particular style of melt...this method allows me to use some older blooms as starting material, thus giving me greater flexibility. Now I will start keeping track of the total carbon added and total carbon not absorbed by the iron to see if skipping the glass is practical. Here are some pictures of the results, the crucible was pulled at full heat and placed in a metal can of ash and sawdust.

 

Jan

Edited by Jan Ysselstein

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...