Ultimate Heat Treat Cell | Part 2

[@Proud2bmodest]

The middle of the part being lapped is always in contact with the lap while the edges are intermittently in contact. Because of this more material is removed from the middle than the edges. The only solution is to use a larger lap or surface plate as a lap.

[@l33tLX]

Exactly this

[@PeterK6502]

Yup, this method is used for creating parabolic mirrors, for the reason you mention.

[@mike-carrigan]

Your laps are smaller than your part so your lap is working the middle of the part as you get near the edge. You can't get it flat unless just your high points are only making contact.

[@jeremyriggs8469]

Once you have a heat differential in the plate (hot knife on one side and chiller pulling heat through it from the back) you'll likely have your aluminum plate pull out of flat anyway. So sweating tenths probably isn't gaining you much/anything.

[@manuelmunch1601]

Hi. In a similar project, I installed a computer radiator with fans in the return. that was a cheap option, both in terms of costs and electricity. I then simply wired the fans parallel to the pump so that they always run when the pump is running. that was really the best way to reduce the return temperature in front of the cooler and to relieve the load on the cooler.

[@tacticalcenter8658]

You want both sides of the blade to touch at the same time. Putting it in the vice the way you are, one side gets cool first and causes warp.

[@GOAP68]

Take a look at the McDonads grills. They made a design some years ago that was oil heated and designed to provide an even temperature across the surface. Had channels like your plates with some additional design features.

[@danmenes3143]

John, what are you smoking? You have a structure made out of aluminum, that is going to be subject, in use, to a temperature gradient of several hundred degrees centigrade per centimeter. You are worrying about tenths? As soon as you put a hot knife blade on this, it is going to go out of flat by the other kind of tenth--something more like 0.1" than 0.0001". FYI, the internet tells me that the tempco of 7075 is about 13 ppm per degree farenheit, or about 25 ppm per degree C. So if your knife is even at 400 C above room temp, that's a growth of 0.01" per inch. That's just the direct linear movement due to the change in temperature. The bow due to the temperature differences will be a lot bigger than that.

[@DudleyToolwright]

The lapping will improve the surface contact and thus lowering the thermal resistance, but getting the surface flat is probably like flattening a piece of playdough. When you press the two plates together and they are separated only by the object in the center(ish) and the plates will readily deform to match the points of pressure - much more than the 0.0001" you were going after.

[@Spirit532]

Just a quick note about pyrometers - you can't directly measure the temperature of metal parts using a pyrometer. Metals, even dull sanded ones, have really really low emissivity and high reflectance in the longwave infrared, which means you're measuring whatever is reflecting in that piece of metal. At the angle you were using the pyrometer, you were just measuring the temperature of the table with a small offset. A simple solution to this is to just cover a piece of whatever you're measuring with black electrical tape and point the pyrometer at that. Most vinyl electrical tape has an emissivity of around ~0.9, which is close enough to 0.95 that most pyrometers are set to.

[@BerndFelsche]

Black acrylic paint spot. Also ensure you measure in the same place. And because it's already there, will heat and cool along with the metal.

[@Spirit532]

@@BerndFelsche Acrylic paint can go very glossy when drying and introduce undue reflections. I'd stick to tape.

[@cruch9]

Curious, are you making sure you are water jetting all parts in the same grain direction on the material? That could cause your issues.

[@ronnydidit]

You should lap those when coolant runs trough. The drastic temperature change might do something as well.

[@DudleyToolwright]

However, there is nothing wrong with going through with the exercise, even if there isn't a practical point. Also, diamond grit comes in much finer grit than 30um - which is approximately equal to 850 grit sand paper. To calculate: 0.030mm/25.4(mm/inch)=0.0012inch AND THEN 1/0.0012inch= 847 bits of grit per inch (if touching). Good surface finish is sub micron.

[@keithviolette5870]

Instead of an arbor press to apply the force, you could use an air cylinder with a precision pressure regulator. If you want to get fancy, you could add a timer to a solenoid valve so the time for each blade cooling off is the same. Also, the other comments about contacting the bottom plate first and cooling the bottom side first might be something you want to investigate. Try cooling the next batch with the other side down first and see if they bow in the opposite direction. With an air cylinder clamp, you could set it up horizontally easily. Hold the part by the foil bag in the gap, without it touching either plate, hit a foot switch to clamp the cylinder so both cooling plates contact each side of the blade at the same time.

[@F0XD1E]

They mentioned a pneumatic or hydraulic press in the last video, but had safety concerns about crushing fingers. Certainly would be a process improvement though.

[@kwk8363]

John, use ultrasonic cleaner for those precision stones. They will be flatter. You will get rid of all the debris from if

[@osfunsiouf]

...could a thick acrylic cover have also given you the visibility to see any issues in the water flow (air bubbles, algae, sealant etc)?

[@paradiselost9946]

as many point out... going way overboard for something that has more flex than the lapping allowance, and will be subject to thermal stress on top of that. anyway, regards lapping... try zinc. flashing. or get them zinc plated/ hot dipped. lead... tin... solder.... plenty of softer metals you can lay on a CI backing plate...

[@PeterK6502]

The lapping technique shown in video doesn't allow for the creation of flat components due to it size relative to the lapping tool; instead, it produces curved surfaces resembling spheres, making it particularly valuable for crafting parabolic mirrors. Secondly, even a perfectly flat part will exhibit a slight bow when suspended on three points. While the exact extent may vary, this phenomenon remains present. This is precisely why parabolic mirrors are upheld by multiple points, serving to counteract this effect.

[@osfunsiouf]

(also, plastic fittings might avoid corrosion better; stainless steel and aluminum don't usually mix well, at least from what I've read)

[@zviper]

i think you need 3 plates to make a perfectly flat plate. Oxtool is the GOAT

[@DeKempster]

Please buy a contact surface thermometer. That IR doesn't work on reflective surfaces.

[@hammurambi]

They do if you set the emissivity of the IR thermometer correctly.

[@DeKempster]

@@hammurambi doubt they did that.

[@prestonengebretson2920]

Hola John, Flipping the Tool and Work from Top to Bottom is a Normal Procedure when we make Optics to keep them either Flat or Keep the Desired Radius...the one On Top wants as you said to Go Concave and thus you want to change them up on a regular basis.

[@omdesigned]

Also seems you'd need steel plates?! The aluminum will bend like crazy.

[@omdesigned]

John, with all your 3d expertise, I’m quite surprised you haven’t designed a ‘daisy’ shaped with the perimeter of the ‘petals’ shaped more or less like your knife blade as seen from above. This could give you a directed quench that hits the entire blade at once from the center and you withdraw heated fluid from the edges. You don’t need a great deal of strength in this area so you can afford thinner walls in favor of greater fluid velocity. This whole design could be imagined as a SpaceX quencher that’s contained within your Al block.

[@Coffreek]

I admit, I haven't watched the whole video yet, so Jon may get to what I'm about to say, in his time. Ox Tools has a pretty detailed video about making lapping plates, somewhere in his archives. In it, he laps three cast-iron plates against each other, and shows his work with an optical comparator. They get very flat. I wonder if something similar could be done with these quench plates? Make them in batches of three, and lap them against each other. It seems to work for same-hardness materials. I understand your concerns about safety, to get as much flatness as you want, you really are going to have to eliminate as much of the "human touch" as you can: pneumatic clamping for consistent pressure, and mechanical positioning so the plates hit at the same time. There are ways to ensure safety with this kind of equipment. Lots of machines have two buttons or paddles along the front edge of the table, and both must be pressed in before the power will turn on. Is it possible that your heat treating foil thickness is introducing some error? It's not made to much tolerance, and may be affecting your ability to keep equal pressure across the blade. Finally, we all know how "Moria" is pronounced. C'mon.

[@steves.3485]

Any reason why you didn’t just lap the two plates together?

[@YouveBeenCabadged]

Stupid question from non-machinist - but how much do we expect the aluminum part to sag under its own weight when it's put up on the 3 points? It looks pretty rigid - but the sag can't be nothing. I'd expect it to be measurable?

[@felixpelletier4296]

"We're at pretty much zero. Within a tenth, I'm just playing around here"

[@DLDB01]

quick thought, could you not just lap the 2 cooling plates on each other?

[@vanguard6937]

With only 2 plates you can induce a convex/concave surface. That's why to get a flat lap by hand, you often need 3 plates.

[@mikeobrien4081]

Lapping and heat treating are cool and everything, but let's talk about that Volvo for a minute.

[@flikflak24]

Have you also considered a constant rotating pitch grinder ?

[@erikslagter3231]

i think the chiller makes a lot of difference.

[@patrick2342]

Use a cooler in place of the bucket.

[@coreyb4073]

Slightly crazy idea, you can put a piece of cellophane tape in the middle, the lap will only make contact on the edges, lap the edges till they are the same level as the center, you will have a high Ring half way to the center then lap the ring flush. Plastic push to connects will distort less, but you should be lapping as a full assembly. Most of my lapping is in diesel fuel injectors so much smaller parts

[@ingentingbra]

Umm. Why didn't you do it on a surface grinder?

[@cruch9]

Your plates may be "flat" but you never checked for parallel between the two faces after installed in press. Constrain the blade in the lower plate with pins so blade is in the middle of plate. Add a pivot in the top plate so it mates with lower plate while pressing.

[@Dustex1]

The top plate has a pivot poin if remember corectly

[@LoneWolfPrecisionLLC]

​@@Dustex1correct

[@tylergilbertson4086]

When you use the drill to torque the bolts down does the drill chuck ever open up and the bit falls out? When doing some of my parts with the same technique to bolt down that was always one of my problems. I switched to one with a 1/4" quick release style like a "DeWalt part number" DCF601F2 and there is no chance of the bit dropping out.

[@davidplatenkamp]

Any reason you're not just running tap water through the plates?

[@billclancy4913]

"This is no lapping matter!"

[@jpiccari]

Heat treating is a dark art and I know less than nothing about it. But why is any of this important? Wouldn’t it be massively more efficient and with a better result to simple quench less carefully then have a quick pass on a surface grinder? Is the aluminum really flat during use if you are lapping it at a different temp? Are the blade really be cooled evenly if the blades are laid on the bottom plate before being pressed? By attempting to prevent warping aren’t you building stresses in the blade rather than allowing a more natural warp then grinding it to final size. I don’t get any of this but I’m also just a humble home machinist so maybe I’m completely missing something.

[@donutfpv]

I could be wrong, but I think the way you're measuring flatness on the 3 points is wrong. If you've got the part on 3 points and you're indicating off of the opposite surface, you're measuring parallelism.

[@jongmassey]

"the proof of the pudding is in the eating"

[@chasesyverud8984]

Does anyone know how long shipping normally takes? I bought a saga yesterday and was just curious

[@steves.3485]

Depending on customs it seems like it takes a week to 10 days to get to the US. Mine unfortunately was stuck in customs for a whole week so it took a lot longer than I wanted.

[@chasesyverud8984]

@@steves.3485 ahhh gotcha. That's the one issue with buying directly from them. But gotta support!

[@titter3648]

I would not use antifreeze in a open bucket like that. The glycol is a death sentence for any shop dogs. It taste sweet to them, and destroy their kidneys.

[@MaDuceRules]

Not sure about Canada but in the US since about 2013 all antifreeze sold has to have a bittering agent in it so kids and animals don’t drink it. However keeping it coved is still a good idea.

[@BerndFelsche]

Isopropyl alcohol can be used as anti-freeze. Should not need much. Water is better for heat removal so the less antifreeze the better. Should not need more than 5% unless you have snowmen on the night shift. Other additives deal with corrosion that might occur between dissimilar metals in the plumbing. Tannin can be used as an antimicrobial if you don't see freezing temperatures. Also, a "sealed" circuit and not one wide open to the air will decrease the growth of gunk. You need just small breather holes to cope with thermal expansion and contraction.

[@roykaltz3233]

Be careful on what coolant your using. It has to be safe for aluminum if the coolant is circulating through the plates. If it’s not, it will eat at the aluminum.

[@F0XD1E]

Plus the stainless steel fittings are going to make a galvanic couple the same way their copper rtv did. At least the coolant should have some corrosion inhibitors.

[@sedled2829]

That plastic thread on the pump got me wincing 😬

[@Daniel-vq9zb]

i bet with some 600 paper and a junk surface plate you could get it within a tenth pretty quick, way over complicating things here

[@davidplatenkamp]

I'm no metrologist or toolmaker, but it seems your part is hydroplaning on the lap because of the tape covering the grooves.

[@paullmight42]

you want your knife maker to be ocd about flatness...John's headstone will inevitably read "It's good, but let's get it better..."

[@richpryor9650]

John, you FOOL!! You're supposed to lap a part on a surface that is larger than it.

[@Stoeckermann]

First? So tempting