Stem ejection mechanism working. F project | EP#9
2024 ж. 16 Сәу.
11 336 Рет қаралды
I am still struggling with development of the stem press, but some progress was made - the ejection mechanism that pushes the glass out of the mold is working!
Thank you for your continuous support!
NEWSLETTER
► www.daliborfarny.com/newsletter
WEBSITE
► www.daliborfarny.com/
INSTAGRAM
► / daliborfarny
FACEBOOK
► / daliborfarnycom
Thank you for taking us along on the process. Exciting insights
Nice progress, great work! Proper respect for not hiding what went wrong, what you learned and how you dealt with problems. No sugar-coating here and I like it.
Nothing as beautiful as freshly scraped ways on a machine.
I saw that too. Did they rework the ways, or is that old thing brand new?
Amazing work! Awsome to see you solve the manufacturing challenges
Thank you! More to come ;-)
6:34 Me : Huh, pins aren't straight. Pins : *italic mode*
Yes, it is much better apparent in the recorded video..
@daliborfarny I'm surprised how you kept your calm when this happened, widhout saynig any bird name in czech😂
I had seen the molds kept warm with a bushy flame between loads to reduce their effective thermal draw on the glass. Eventually the molds will heat up enough that the melting flames would keep them at temperature. also be sure the bottom of the exhaust tubeulation is supported underneath or it's weight will sag it down and away and constrict or just seal off just under the wafer even after the mandrel on the upper die is withdrawn. They used to blow a puff of air up through the hole to make sure that glass sets pretty quickly but not enough to set up stress in the glass as it cools. Looks good. I am half way through rebuilding the Eisler single head pinch flat press stem machine and have already made some of my new design stems on it. Working on an easier method of flame control and better quality torches this week.
I did like 20 attempts and the molds got so hot the glass started sticking (that happens over 350C). We will see if SS makes difference! Good info with the air blowing - I hope I can avoid it - adding that to the rotating carusel would be hell :-)
@@daliborfarny You need to try different materials, different cast irons, cast irons with vermicular graphite...
@@daliborfarny I mean, I'm just getting this from google, but Graphite doesn't stick to glass, so maybe look for a cast iron that has a stupid amount of carbon / graphite impurities... May need a graphite powdering every so often.
Столько лет упорства в восстановлении и развитии этой древней технологии! Моё восхищение!!!
@@bipin2630 Дуровы и Маски нервно курят в коридоре!
You're doing amazing work. This is a perfect nexus of machining youtube, glass blowing, and vintage tube tech!
Krůček po krůčku se blížíte ke konečné dokonalosti, držím palce Dalibore a díky za super video 👍
🙏 dekuji! Jeste to chvilku zabere :-)
I love watching your progress and learning it. When pressing pins always push them a bit and then back the pressure clear off, that will allow them to recenter, and press again.
Why not have a premold step for the glass slug that makes it closer to the shape you're looking for so you don't have to heat the contact pins as long
I wouldnt have thought taht it is so difficult to reinvent the wheel. Keep on !
This project is so awesome I'm sure many engineers from around the world wanna be a part of it. I know I am 🤩
Thanks for taking the time and effort of making these fantastic videos. I find it very interesting to see the problems You face and how You work Your way to viable solutions :) Best regargds
I haven't finished the video, but at 2:19 , the "lead weights" likely refers weights made of the metal: lead (i.e. Pb) - not "lead" as in a "wire lead". Lead weights were extremely common (at least here in the U.S.) and were used as fishing weights and for builder's "plumb bobs", etc. RCA using weights made of lead metal as part of a manufacturing process would make sense - especially in that era (1950s).
Nice work! I think it might be worth experimenting with a tightly toleranced two part mold at the top, such that the diameter of the mould that makes the OD of the glass part slides down first, without pressing on the glass, then the top face comes down and presses the glass to final thickness / squeezes the glass out to the diameter. The key thing will be to ensure all the meal parts are at a close, and stable temperature.
Making real progress! Looking good
Can you use an alloy with a higher melting point than the correct softening point of borosilicate? Also use torches that you can aim so the bulk of the flame is next to the circle the pins are in instead of on the pins and the glass. There are positionable lathe burners with multiple heads but I'm sure you know this
You should take a few design queues from the old videos that still exist of the original process... In particular, the placement of multiple stationary smaller torches are spread out in multiple even places surrounding the parts, that way the glass is more evenly melted, helping reduce uneven temperature, and lopsided collapsing. This also allows you to direct the heat directly at the glass, and not the pins, this should give you the ability to introduce more heat into the glass faster, and reduce the chances of you melting pins. As for the center flanged tube, you can easily shape the glass at that point as well by putting a nipple on your upper die that slips into the tube of the flange, and helps form the glass at that location, while maintaining the opening.
Absolutely fascinating!
maybe you've already tried it, but i was wandering if there is a way to insert the pins into the molten glass simultaneously with the press-forming, so they won't overheat.
This is a good idea. You might have problems with the pins becoming stuck in the dies without lubrication and the tight tolerances involved to keep the pins from becoming encapsulated.
Great progress!
Very interesting. I admire your persistence.
Glass seems tricky to work with. I hope you can rediscover the process!
I was going to mention the cold top die and pulling the heat away.... but, you covered it at the end talking about stainless instead of cast. But still wonder if the top die being heated would also limit the amount of heat pulled from glass.
Slowly making progress, can't wait for the first ones to go on sale. Started working on the electronics and software to drive them, using small Russian tubes for the prototype.
I hope you get these made!
I wonder if you could use a CO2 laser to get more precise heating of the areas you want without overheating the pins.
LLC Melz is amazing, I got some sample electron guns for CRT's from them.
A shame it's a Russian import, that will make it very difficult to get right now for most people. I have to wonder if anyone makes them in North America, there is still industrial production of photomultiplier tubes iirc.
@arfink the import situation sucks for sure, but there isn't anyone else that I've found making this stuff. Hopefully things will improve with time.
Very cool project!! Love the toolings you are making yourself. Also, wouldn't you want special sealing alloys that would form a chemical bond with the glass at the sealing interface as well as matching the thermal expansion of the glass? At least I think that's how it's traditionally done
Stainless will work, and the oxide layer will also help. Anneal the stainless before final shaping, it will be work hard, so anneal before machining, and again before the final grind.
Interesting, I will have to look that up!
@@daliborfarny Careful machining it, it needs coolant and sharp tooling as well, or it turns into a hard mess on the second pass.
I'm amazed that you can melt tungsten pins.
Question, if you make the fill / flanged stem nearly to the end size and while hot stamp/punch holes where pins would go. Once that is in your stamping jig, you would require only a small glass ring to fill around the pins, since the stem / inner part is already formed and in place.
Lead also has a low heat conductivity, so it's less likely to make the glass crack from rapid cooling. Maybe you can use magnesium oxide as alternative. They sell some easy to carve stones for artists and crystal lamp crap. I think they call it soap stone in english. Best of luck.
If you do go stainless you can get it electropolished it's normally pretty cheap and would be able to do all the small details of your moulds. Cutting on your small CNC could be interesting. Low depth of cut and keep the feed rate up, rubbing the tool causes work hardening and that's bad. I do wonder if you could include voids in your moulds as somewhere for excess glass to go? Just to make getting the fill exactly right not as critical.
What is the alloy of the pins? I'm engineering something that has pins through glass and your pins seem to work good
hello ! for finishing the parts, use a tumbler ! the surface condition is perfect
Will give it a try!
@@daliborfarny exemple : turbo twin tumbler with every sand blast or tumbler rotary with stainless steel needle
@@daliborfarny bead blasting might give you an interesting surface texture too. Each way bet if it'd be good or bad.
I don't know if you could, but a DLC coating would help thermal conductivity might not have to change material. i wonder how alcohol blow lamps would fair might save a lot of money on gas I cant wait to see what you do next
Awesome work as usual. 👍 Why didn't you put the little pins in the freezer and/or heated the other part? That way you would have a bit more clearance to press them in easier 😊
They are designed for a serious press fit, I dont think heating it up would help.. anyway, I hoped I found a quick/simple solution for inserting pin into metal.. wasnt that easy in the end.. :-)
As for the glass extruding outside the ring, could you make the outside step taller so the die closes before the pressing action starts? that way the glass wouldn't have anywhere to go but towards the centre yeah, stainless should help a lot with thermals, might be worth looking into high temp grades for final dies, like 321 or INCO 825, or even SAF2507, Inco can be a bit expensive, and all 3 are a pain to machine, but they should last years working at glass working temperatures. btw, most high temp Stainless grades have even lower heat conductivity than basic 304/316 grades.
Mullard had an interesting technique for the base that can be seen here: kzhead.info/sun/aLt8h8minISIkqs/bejne.html They put a glass ring only around the outside of the wires/pins. They then heat the ring and use a toothed wheel (presumably to clear and go between the pins), which pushes some of the outer ring through the pins and towards the center. Then, they press the base. Using the toothed wheel apparently removes the need for an inner glass ring to form the center. Also, the outer ring can be heated directly with a simpler torch setup. Anyway, I thought this video may provide some ideas. Good luck!
Could you use powdered glass instead of the rings? And maybe inductively heat the mold enough to melt it?
Have you considered putting the pins in the top part of the mould and once the glass is molten in the bottom part you press the pins trough it. It could help heating the glass without melting the pins.
You probably want an ejection mechanism from the bottom that presses on the pins to extract the stem, why don't you partially lift it between the presses to heat it. I'm not sure if a different moulding material would help much.
The facebook video is what i was thinking of recommending on doing too, the center needs to be a plate. Also calculate the volume of glass required and add like 10% extra. As for the mold if you want to prevent flushing on the outside you can design it so that it forms a seal from the outside to inside as it's lowered (aka different draft angle); another way is to add a flushing groove for overflow with a thin passage so any flushing breaks off cleanly from the outside. I am not any glass expert but i am mech eng, and this is what i would try without any research and without knowing the requirements;p
I work with glass to metal seals as my main job, and we still have a base press machine for certain jobs What metal do you use for the mold? Most of ours are made out of Monel. We also use beeswax on the top jig to prevent it from sticking, so if you still have problems use beeswax for lubrication. Really cheap, does not affect the final product at all, might be slightly harder to do for a fully automated machine but really useful. For best practice the pins should be beaded up with glass before adding on the larger rings. This can be done in 2 ways - Having pin sleeves (Thin tubing with internal diameter close to size of the metal pin), and working them on the base press by heating and allowing the glass to gather, or by individually beading up the pins on the lathe with glass rod. Rod's cheaper and easier to come by but the pin sleeve method is slightly easier and almost guarantees correct position. Having the glass on pins before pressing drastically reduces the risk of a poor bond, and of bubbles between the pins and glass (a possible leak path for when it's pumped under vacuum). Not every press has this, but ours has a lift on the bottom jig. This allows us to lift up the piece while it's still warm (first press will cool it enough to be fairly solid) and then get the heat from the ring burner underneath the piece to better heat the centre. Having a flame underneath allows the heat to spread and rise better. We also have a hand torch that we point downwards towards the centre for certain jobs (and is what we'd use if manufacturing your setup). Outer and inner ring of glass makes more sense than trying to squash a larger amount of glass into the centre, and especially as there's no inner ring of pins that flared piece of tubing inserted into the bottom jig would also be useful, you just have to experiment with how much glass is on the inner rings. We also have a small carbon tool to hand (flat paddle) to tool in the outer glass slightly between presses so it can fit in the mold without it spreading out too far. Try 2 or 3 softer presses, don't try to get the entire shape at once. Glass has memory to it and doesn't like being forced into a drastically different shape, no matter how hot it gets. Even with casting in carbon on an induction coil, we will often run an item a few times before we're happy with the result. Slow and steady wins the race.
That has been my experience too. Multiple quick heatings and pressings unless its on an automatic machine which usually have a pre forming position with a little paddle that comes in pushing the glass the direction you want it to go in. Also automatic machines will deposit a very melted nearly fluid blob of glass in the middle. Most places I went to had foot operated push up mechanisms for the wafer or button stems too. They would push by the pins as I recall. We only made flat press and preform type stems in house. I discussed the idea of pre beading the pins with Dalibor. With Iron nickel pins and AR glass it really wants to be done in one operation as the degree of oxide diffusion is pretty great and multiple heatings could result in Leakey or weak seals. Think of it as borated dumet. They want only a single heat or you will loose the oxide and give you a bright seal which is weak. Which company do you work for? Fredricks, Richland, Or Troyonic by any chance? Those are the places we used to work with that still are around.
@@MrSupro nah none of them, Torr Scientific, which is a "new" company, but if you've heard of ITL back in the day most of the equipment and like half the staff are from there. UK based
Oh you make the fancy stuff. The only Uk company we ever talked to for glass was Moore’s but they were too expensive for us.
Maybe thermal capacity of the top part is big and causes instant freezing. I think that original machine uses less cast iron and more graphite parts that are not visible. Maybe gentle preheating of top part will make better parts. I am watching the whole series as it is good mental workout in design and engineering. Hope to see new video with goot news soon.
I am afraid this may play a role too.. graphite is close to SS with thermal conductivity (24 W/mK), it would work well in the upper mold, we will see! Thanks for watching!
@@daliborfarnyAt work our customers use white cast iron with deeper cooling channels to prevent overheating of moulds in our IS machines in the glass bottle industry. There is a preheat time involved before steady state temperature is reached during production. Graphite and ceramics are used when pre annealed bottles require handling when going into the oven but graphite is generally too soft of a material for our application if used as a mould.
Thank you for these interesting & frequent updates! Would it be possible in a first step to "punch" the pins through the flange (which already has the final dimension/circumference) instead of "flowing" the glass around it and in a second step build the stems around the pins by adding only a small glass ring if needed? If the starting flange is thick enough, could there even be enough material to do it all in one step?
put a small diameter glass tube on each metal lead. just a short length of tube maybe 5mm long.
Is it an option to flip the unit, and have the pins on the upper half of the mould? That would mean you wouldn’t have to worry about overheating the pins and could use a glass disc instead of a tube.
Would it be too costly to make a pre formed puck with holes for the pins, then press that? It seems like most of the problems are related to glass flowing where it isn't supposed to be or is missing from, so making the distance glass has to travel would be logical solution, no?
I lke that Your progress is definitely in the positive direction. In looking at the evolution of the inner melt perhaps you could replace the inner cylinder with a stem that has a glass mass that is expanded to about the size of the ring when the outer cylinder goes molten. It seems to me that this might make the mass of the central glass more controllable thermally and keep the stem clear. Still not certain how you're going to control the differential thermal shock on the stem without a controlled heating of the lower mold. Perhaps you could use a PID-controlled induction heater to bring the lower mold to bring it to a thermal state just below the liquidus state of the molten glass. Relearning this level of historical materials science isn't easy or anyone could do it..
Make sense, I still hope to use only glass rings as it makes things easier. The thermal shock/stress is not a problem if you can extract the stem still hot from the molds and move it quickly to annealing oven to release the stress. The mold will achieve some operating temperature after few minutes. It needs to stay below 300C, otherwise the glass starts sticking to it..
@@daliborfarny It seems like you have a handle on it. I think your point on where the burners point is one of the keys to controlling the melt so you don't underheat or overheat the molten mass. Still not sure how you're going to control the mass in the middle if you're not going to use a preformed stem. I still think that, if you're going with a stem then the preformed flare should contain containing all of the glass that goes inside the pin ring. That's only a grain of salt observation in the absence of other criteria. Have fun!
As I said in my comment ts last time. You don't have a hope of flowing glass to the centre unless you preheat the mould itself. Glass is rapidly cooling below it's transition temperature the moment it touches the steel.. In that company video the lower mould is preheated.
There are 2 words in English with same spelling "Lead" one pronounced led, which is metal, .and one pronounced " leed" which means to show the way, .so Lead Weight was referring to heavy metal
great progress! you'll get there! You can borrow some Philips burners if you need , I want to copy them when machines are in the new workshop.
Thanks, I have some spare ones!
Why don't use a glass disk with perforated holes for the pins, so you can heat up the superfice
Hey, how about building induction heating into the moulds to heat them up?
I really think you would benefit from root cause analysis. That would reduce your T/S time. It seems like your are using a brute force model which does work in a low volume/cost sphere but expensive and time consuming for you.
would it work if the flange of the exhaust pipe is big enough to cover the inner area so no inner ring is needed and only the external ring needs to be melted?
Yes, thats what Melz uses in their video, might be the way..
It would still need to be brought up to seal temperature or a radial thermal stress fracture would occur. This is however a method that has been used where a exhaust tube and flange of a calculated volume is molded in a separate automated process. Alternate to this is a molding process where the exhaust tube and wafer are pressed with holes and pre beaded pins are installed in a subsequent step. The fastest over all though for semi automatic production is the process seen in the Melz video. They have an automatic cutting and flaring machine making the flanged tubes.
Why use these weak pins when there is ground rod already available?
Maybe they ment LEAD , like Pb , weight ... or am i confused ? :)
But Pb would melt right away in this environment... I am contradicting myself now or i am missing some detailes :)
Lead as in the Pb element, does not make sense in that context. They are really referring to the pins as leads (conductors).
But that would mean “heavy weight” which has no reference to anything.. still think it is the pin :-)
@@daliborfarny Lead has always been used as the term for the metal pin since the edison days here in the US. and england.
@MrSupro agreed. Pronounced differently than the element. Lead, as in, lead a horse to water... meaning "connection point termination". The test cords of your test equipment are also "leads".
is it maybe possible to pierce the pins through a molten glass donut ?
The Melz torches look like methane/air, not oxygen. Lower flame temperature might help to not melt the pins. Are you using propane/oxygen?
Yes, I use propane/oxygen. Methane/air would be too cold I guess, but they have 6 nozzles..
Could the pins start out in the top die. Then you melt the glass in the lower die and the press the pins through the molten glass below?
F stands for F-ficiency and F-fordable!
Maybe they mean the metal lead. (Element 82) It's unlikely to stick to glass.
How does setting up super chats or channel memberships? I don't use patreon but I'm able to help some.
Looks like the pins used by MELZ are a much higher melting point metal? Would it be possible to use tungsten pins?
My guess is that is a kovar/hard glass type stem in which the temperatures will be quite a bit higher. Here soft glass and nickel/iron leads are being used. Generally kovar seals are made at bright orange heat and soft glass seals at bright red heat. Even though Kovar and alloy 42 for example have similar melting points the alloy in kovar is more oxidation resistant at temperature than the nickel iron alloy of the alloy 42. Its not so much melting as it is oxidizing and burning away. Another thing to remember. Soft glasses are nearly always worked with a highly oxidizing flame or you will discolor/cloud the glass. Essentially keeping the glass in oxide form. Unfortunately the metal pins are also trying to turn back into oxide as well in those conditions. As for Tungsten, yes it could be used, but the cost of the pins and appropriate glass would be unaffordable for this product. Typical Tungsten type sealing glasses are as follow. Schott 8487 is very expensive and only occasionally available with very large minimum quantities are demanded while corning 7720 nonex glass is lead bearing and long out of production. I work with them because I have left over stocks from closed companies to play with. From a new product point of view a bulb and stem set even made in house for this size product may have a $50 to $100 material cost in quantities of thousands of pieces and that does not include the labor or loss. Fine for a multi thousand dollar military thing but not for a retail nixie clock.
They are the same alloy. Tungsten would need different glass with much higher melting point..
Man, I really enjoy all the process of learning and improving your products and manufacturing you're going throght for a quite long time. I follow you since your first videos and wanna buy a few of them to make a clock, but unfortunatelly I cannot afford them right now. So I really hope you can make more affordable models. I wish great success for you and your business!
Hey, thanks for the support, thats what I am trying to do here! :-)
Great work. I love that ultrasonic mould polishing tool. I often wonder what sorts of exotic tools that die makers have to polish intricate cavities inside the die forms.
A stupid question - couldn't it be possible to push pins after the glass is melted?
That would be probably even more complicated - I go the way the old factories went..
It seems like you should just replicate the existing manufacturing process instead of trying to troubleshoot. Use the flared stem plus one ring of glass on the outside.
Perhaps insert a thermoelectric cooling device under the lower mold half, with the idea being to wick heat away from the pins? Or would that prevent glass sealing around the pin? Really neat stuff. 👍
That ultrasonic mold brush is really neat!
When the pins folded - Noooooo lol
so sexy 🤤