A quick update.
Thanks to MarkRob's help with this line of code, the lpiValue calculation works now and I have a valid LPI number being displayed! Thanks Mark!!!!!
lpiValue = ((float)1777777.77777778 / (float)count)+0.5; //Calculate LPI based on the y=mx+B equation for 1/F vs LPI and round up.
Lead in = 16LPI (but adjustable)
Lead out = 4LPI (but adjustable)
Cut LPI is adjustable.
I ended up going with an Arduino Nano so I could fit things in this 1U box. A few more things need to be tweaked inside the box, then start wiring the back panel. On the front are adjustments for lead in, out, and cut LPI and a switch for Remote / Local LPI adjustment. For now, the front panel sets the LPI. When the back panel connector arrives then I can use the remote switch and a lathe mounted LPI pot.
I'll also be adding MIDI at some point so I'm installing those connectors as well.
One thing I'm changing tonight is making the ramp time adjustable. Right now it's fixed and may need to be a bit quicker.
That's the update for tonight.
Bryan
Bryan's Fairchild 539A lathe project
Moderators: piaptk, tragwag, Steve E., Aussie0zborn
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
The box is done - except for the rear connection for the remote (lathe) mounted LPI input pot.
I ended up putting 2 top / rear panel disconnects for the Touch screen, 12V switched LED lighting and the DB25 connector to the micro-stepping drive. That makes it a bit easier to remove the top / back panel for mods.
MIDI in/out jacks are not connected or implemented yet.
The touchscreen is mounted with Velcro on the lathe, and the "remote" LPI pot fits an existing hole next to the power switch.
So far, the whole project uses only existing holes. The wires for the LCD panel also drop into an existing hole where it is currently located.
That's it for tonight.
B
I ended up putting 2 top / rear panel disconnects for the Touch screen, 12V switched LED lighting and the DB25 connector to the micro-stepping drive. That makes it a bit easier to remove the top / back panel for mods.
MIDI in/out jacks are not connected or implemented yet.
The touchscreen is mounted with Velcro on the lathe, and the "remote" LPI pot fits an existing hole next to the power switch.
So far, the whole project uses only existing holes. The wires for the LCD panel also drop into an existing hole where it is currently located.
That's it for tonight.
B
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Hey guys,
Lots of progress. Thought I would show you a few things.
The control box is working awesome! And I'm very happy with the ramping rate between speeds (LI, LO, cut) . I decided to do a quick test cut on a nasty 1947 16" disc and take a look at the groove spacing. I used an old school machinist ruler that has marks every 1/64th inch and took some shots with my microscope.
First up, Lead In (set for 16LPI) and displayed as such on the LCD display:
Note that there is a groove every 4 marks on the scale. That means the space between the grooves is 4/64" or 1/16". That means there are 16 grooves per inch - or 16LPI! Yay!
Next up, Cutting. Set for 192 LPI as shown on the LCD display:
I picked 192 LPI since it's an product of 64 for ease of measurement.
As you see, we have three grooves for each line. So for each 1/64th inch mark, we have 3x the number of lines - equating to 192 LPI! Awesome!
And lastly, Lead Out. This was set for 4 LPI.
Is it? You bet!
As you see, the groove spacing is one line each 16/64" or 1/4". Over an inch, that's 4LPI.
Yep.
For all the pictures, I tried to use "40" on the scale for the reference...
More stuff to come in the next post...
Bryan
Lots of progress. Thought I would show you a few things.
The control box is working awesome! And I'm very happy with the ramping rate between speeds (LI, LO, cut) . I decided to do a quick test cut on a nasty 1947 16" disc and take a look at the groove spacing. I used an old school machinist ruler that has marks every 1/64th inch and took some shots with my microscope.
First up, Lead In (set for 16LPI) and displayed as such on the LCD display:
Note that there is a groove every 4 marks on the scale. That means the space between the grooves is 4/64" or 1/16". That means there are 16 grooves per inch - or 16LPI! Yay!
Next up, Cutting. Set for 192 LPI as shown on the LCD display:
I picked 192 LPI since it's an product of 64 for ease of measurement.
As you see, we have three grooves for each line. So for each 1/64th inch mark, we have 3x the number of lines - equating to 192 LPI! Awesome!
And lastly, Lead Out. This was set for 4 LPI.
Is it? You bet!
As you see, the groove spacing is one line each 16/64" or 1/4". Over an inch, that's 4LPI.
Yep.
For all the pictures, I tried to use "40" on the scale for the reference...
More stuff to come in the next post...
Bryan
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Next up, some mods to the overhead...
Although the AirPot and clamp on bracket set up shown early on seamed to work in a static "tap and see response way", it really didn't work when cutting. At first, I thought the issue may have been a lack of stiffness with my plastic clamp on scheme or the thin aluminum bracket, or perhaps because the dampener wasn't attached near the head. Anyway, I decided to come up with something that would solidly attached to the carriage and that other things could also be attached to. I'll go into that in another post.
I didn't want to drill and tap any holes. And this section on the back of the carriage looked like a good choice. There was one existing threaded hole, and what looked like a flat area where a plate could be attached, so I made a plate and drilled holes to mount something (shown later).
It turned out the plate wasn't stable, and what looked like a flat surface was no where near flat on the carriage. So I took a diamond hone and knocked off the high spots so the next assembly would sit flat. Here is the plate, and a spacer that turned out to be needed so the next part would clear the gear box housing on the left, and a section of V-Slot aluminum extrusion.
There was only one very small screw and threaded hole available, so before I attached the entire unit I placed a very thin layer of RTV on the flat part of the carriage, let it set up for a while and placed the unit in place and lightly tightened the screw. Later, I snugged the screw... The damn this is rock solid. And if for any reason I need to detach it will be possible because of the RTV.
Here is what the mounted assembly looks like:
I know, some people are going to think WTF on V-Slot stuff...
Well, that will be more clear on the next post.
B
Although the AirPot and clamp on bracket set up shown early on seamed to work in a static "tap and see response way", it really didn't work when cutting. At first, I thought the issue may have been a lack of stiffness with my plastic clamp on scheme or the thin aluminum bracket, or perhaps because the dampener wasn't attached near the head. Anyway, I decided to come up with something that would solidly attached to the carriage and that other things could also be attached to. I'll go into that in another post.
I didn't want to drill and tap any holes. And this section on the back of the carriage looked like a good choice. There was one existing threaded hole, and what looked like a flat area where a plate could be attached, so I made a plate and drilled holes to mount something (shown later).
It turned out the plate wasn't stable, and what looked like a flat surface was no where near flat on the carriage. So I took a diamond hone and knocked off the high spots so the next assembly would sit flat. Here is the plate, and a spacer that turned out to be needed so the next part would clear the gear box housing on the left, and a section of V-Slot aluminum extrusion.
There was only one very small screw and threaded hole available, so before I attached the entire unit I placed a very thin layer of RTV on the flat part of the carriage, let it set up for a while and placed the unit in place and lightly tightened the screw. Later, I snugged the screw... The damn this is rock solid. And if for any reason I need to detach it will be possible because of the RTV.
Here is what the mounted assembly looks like:
I know, some people are going to think WTF on V-Slot stuff...
Well, that will be more clear on the next post.
B
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Hey guys, thought I would do an update.
Lot's of things going on with this project and more updates to come.
But tonight I thought I would share the root cause of an issue I've been chasing.
Everything I have been doing the last few weeks have made things better, but today I identified the primary issue.
The inner groove pattern was cut today. The outer was cut back in the 1940's on this junk disc.
What you see are light and dark bands which result from vertical motion in a silent groove.
Eventually, I counted the bands on the disc and a light went on in my brain... There are 54 cycles of this pattern per rev at 33 1/3 rpm. I looked at the manual for the Fairchild and read this:
So, the 54 per rev pattern on the disc had something to do with the input to the gear box - and the motor driving it.
Several weeks ago, I shot slow motion video of my air based vertical dampener as I was setting it up with "tap testing" and making adjustments. I also shot some video of when the platter drive motor starts up. What I saw on motor start up was a rather violent jump of the motor and quite a bit of vertical movement. I figured I would look into that - and I did.... That's where it got interesting today.
Here is the coupler between the motor and gear box:
I pulled the coupler out, and discovered the motor and gear box shafts were significantly miss-aligned. The motor was way too low and had been that way for ever - perhaps 60 years.
The coupler is 3/8" in each end, so I slipped in over the end of a drill bit and found one end was misaligned by over 1mm, using a drill bit as a feeler gauge as seen below:
More on the next page:
Lot's of things going on with this project and more updates to come.
But tonight I thought I would share the root cause of an issue I've been chasing.
Everything I have been doing the last few weeks have made things better, but today I identified the primary issue.
The inner groove pattern was cut today. The outer was cut back in the 1940's on this junk disc.
What you see are light and dark bands which result from vertical motion in a silent groove.
Eventually, I counted the bands on the disc and a light went on in my brain... There are 54 cycles of this pattern per rev at 33 1/3 rpm. I looked at the manual for the Fairchild and read this:
So, the 54 per rev pattern on the disc had something to do with the input to the gear box - and the motor driving it.
Several weeks ago, I shot slow motion video of my air based vertical dampener as I was setting it up with "tap testing" and making adjustments. I also shot some video of when the platter drive motor starts up. What I saw on motor start up was a rather violent jump of the motor and quite a bit of vertical movement. I figured I would look into that - and I did.... That's where it got interesting today.
Here is the coupler between the motor and gear box:
I pulled the coupler out, and discovered the motor and gear box shafts were significantly miss-aligned. The motor was way too low and had been that way for ever - perhaps 60 years.
The coupler is 3/8" in each end, so I slipped in over the end of a drill bit and found one end was misaligned by over 1mm, using a drill bit as a feeler gauge as seen below:
More on the next page:
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
I spent a bunch of time adjusting the motor carriage that is suspended by springs to get the motor shaft and input shaft to the transmission exactly inline. I used bubble levels to set the lathe to level, then used levels on the suspended motor carriage to level it. Then used a level and feeler gauges get the two shafts aligned... Change one thing, and everything changes.... It was worth it. I'm convinced everything is aligned.
I massaged the coupler, knowing it was misaligned and thought I might work it back into a free state for a quick test. Not so good. I set up a dial indicator as shown below. Because the coupler is damaged and misaligned, it resulted in 0.025" ( 0.635mm) of vertical displacement of the motor each time it rotated. See below: I have several new couplers on order and will try them out if needed. But I decided to measure what I have and tear things down. Temporally, I mounted the internals to a 3/8" bolt and stripped it down. I got a chunk of Latex to play with locally (the guys from the local rubber hose shop thinks it's latex) and have some on order. I'll also order a precision ground 3/8" shaft to assemble this if needed...
Right now, the parts are sloshing around in a Xylene solution and it will hopefully get this stuff off the aluminum. Then, I'll do a quick test fit with the tubing I have and see how it goes. If I like what I see, I can use glue to make the fit (semi) permanent.
That's it for tonight.
B
I massaged the coupler, knowing it was misaligned and thought I might work it back into a free state for a quick test. Not so good. I set up a dial indicator as shown below. Because the coupler is damaged and misaligned, it resulted in 0.025" ( 0.635mm) of vertical displacement of the motor each time it rotated. See below: I have several new couplers on order and will try them out if needed. But I decided to measure what I have and tear things down. Temporally, I mounted the internals to a 3/8" bolt and stripped it down. I got a chunk of Latex to play with locally (the guys from the local rubber hose shop thinks it's latex) and have some on order. I'll also order a precision ground 3/8" shaft to assemble this if needed...
Right now, the parts are sloshing around in a Xylene solution and it will hopefully get this stuff off the aluminum. Then, I'll do a quick test fit with the tubing I have and see how it goes. If I like what I see, I can use glue to make the fit (semi) permanent.
That's it for tonight.
B
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Wow, great way to start the day!
Xylene did the trick and the aluminum parts cleaned right up!
Both ends of the coupler slide on and off this 3/8" drill bit shank with zero misalignment!
And the fit of the latex is so tight I'm not going to bother gluing it. It turns out the aluminum parts have a bunch of gripper grooves!
Now comes the fun part... Fingers crossed.
B
Xylene did the trick and the aluminum parts cleaned right up!
Both ends of the coupler slide on and off this 3/8" drill bit shank with zero misalignment!
And the fit of the latex is so tight I'm not going to bother gluing it. It turns out the aluminum parts have a bunch of gripper grooves!
Now comes the fun part... Fingers crossed.
B
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Progress...
Yes, the latex had to be glued. I ended up using rubber cement and let it set up for a few hours.
Now, get this... Damn near zero vertical misalignment! Notice position of coupler in each photo. The neadle didn't move through a full rotation of the flywheel!
Now, the cut test using the same disc as earlier. The mid band was with the misaligned coupler, the inner band with the rebuilt coupler.
If you look closely, and at the correct angle, you can still see vertical patterns if you know they are there. The motor does vibrate slightly even when there is no coupler installed and its running. My guess is it's out of balance. It turns very smoothly by hand. I'm not sure what the ideal solution is but will try a few things short of finding someone who can rebuild and balance it.
I'm going to try installing 4 oil based dampeners on the motor suspension, one for each of the springs. I've got the scheme worked out, and have 2 to start playing with (the other 2 are on order). If it's a complete fail, I may end up belt driving the transmission with a servo. That would give me the ability to run at half speed like on my Presto. Hummm. This could be an improvement?
That brings up the other topic:
I decided to set aside my head air dampener for now, and I started playing with an oil dampener. Since I was trying to chase down the whole vertical modulation issue and the air dampener wasn't fixing things (wonder why - Ha!) Let me show you a quick photo of what I'm up to now...
I'll go into these dampeners in more detail later, but this one is a modified / stripped down shock absorber for an RC car. The coil has been eliminated. The upper seals were eliminated, the top was milled off for more shaft clearance, the disc with 4 bypass fluid holes (normally attached with "C" clips) was glued to the shaft and on and on. Again, more details later.
Anyway, the center band and inner band on the disc both used this dampener with 25 weight silicon oil. Later today I'm going to purge the 25 weight and try 100 weight. I had already used 100 weight way back before I figured out the issue had to do with the motor so it is worth looking at again.
OK, time to get back to the project. More to come...
B
Yes, the latex had to be glued. I ended up using rubber cement and let it set up for a few hours.
Now, get this... Damn near zero vertical misalignment! Notice position of coupler in each photo. The neadle didn't move through a full rotation of the flywheel!
Now, the cut test using the same disc as earlier. The mid band was with the misaligned coupler, the inner band with the rebuilt coupler.
If you look closely, and at the correct angle, you can still see vertical patterns if you know they are there. The motor does vibrate slightly even when there is no coupler installed and its running. My guess is it's out of balance. It turns very smoothly by hand. I'm not sure what the ideal solution is but will try a few things short of finding someone who can rebuild and balance it.
I'm going to try installing 4 oil based dampeners on the motor suspension, one for each of the springs. I've got the scheme worked out, and have 2 to start playing with (the other 2 are on order). If it's a complete fail, I may end up belt driving the transmission with a servo. That would give me the ability to run at half speed like on my Presto. Hummm. This could be an improvement?
That brings up the other topic:
I decided to set aside my head air dampener for now, and I started playing with an oil dampener. Since I was trying to chase down the whole vertical modulation issue and the air dampener wasn't fixing things (wonder why - Ha!) Let me show you a quick photo of what I'm up to now...
I'll go into these dampeners in more detail later, but this one is a modified / stripped down shock absorber for an RC car. The coil has been eliminated. The upper seals were eliminated, the top was milled off for more shaft clearance, the disc with 4 bypass fluid holes (normally attached with "C" clips) was glued to the shaft and on and on. Again, more details later.
Anyway, the center band and inner band on the disc both used this dampener with 25 weight silicon oil. Later today I'm going to purge the 25 weight and try 100 weight. I had already used 100 weight way back before I figured out the issue had to do with the motor so it is worth looking at again.
OK, time to get back to the project. More to come...
B
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Crap!
Fired things up for the next test and the motor side of the coupler detached.
So the the aluminum parts are back into the "dip" to strip the contact cement.
I'm thinking I will try super glue tomorrow night... That stuff bonds EVERYTHING!
"If it's not one thing, it's something else."
B
Fired things up for the next test and the motor side of the coupler detached.
So the the aluminum parts are back into the "dip" to strip the contact cement.
I'm thinking I will try super glue tomorrow night... That stuff bonds EVERYTHING!
"If it's not one thing, it's something else."
B
You do not have the required permissions to view the files attached to this post.
Re: Bryan's Fairchild 539A lathe project
This is awesome. Explains a couple of reasons why I struggled using arduino for lead in/lead out tracks - Next time I'll be trying to ramp up/down.
Out of curiosity, is it possible to bypass the relays and instead use a PWM output of the arduino to fool the voltage to frequency converter? I'm guessing you'd only be able to go 0-5v instead of 0-10v so you'd lose some accuracy?
I've been told the clockspeed on Arduino's aren't too accurate either - I'll be finding out when I start my Arduino controlled turntable later this month..
I take it the current path is Arduino -> Relays -> Analog Voltage Outputs -> Frequency Converter -> Stepper Driver?
Out of curiosity, is it possible to bypass the relays and instead use a PWM output of the arduino to fool the voltage to frequency converter? I'm guessing you'd only be able to go 0-5v instead of 0-10v so you'd lose some accuracy?
I've been told the clockspeed on Arduino's aren't too accurate either - I'll be finding out when I start my Arduino controlled turntable later this month..
I take it the current path is Arduino -> Relays -> Analog Voltage Outputs -> Frequency Converter -> Stepper Driver?
Record Lathe Embossing Supplies - http://www.supplies.johnnyelectric.co.nz/
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
It might be, but you will need to look into the resolution of the PWM on the arduino. I think they default to 8 bits. Perhaps they go as high as 12 bits with some trickery. Then you need a filter to get DC. As far as 0-5V vs 0-10V goes, you could always throw an opamp in there to scale it for the V/F converter. Depends on what frequency range you need out of the V/F for your application.jjwharris wrote: is it possible to bypass the relays and instead use a PWM output of the arduino
Yep, it's pretty much like this: The arduino doesn't do much but run the display, respond to the touch screen and control the relays. Except it also reads the V/F output frequency and converts it to LPI, knowing that it's running at 33 1/3 prm. That Fairchild can only run at 33 or 78 and I'll never use 78.jjwharris wrote:I take it the current path is Arduino -> Relays -> Analog Voltage Outputs -> Frequency Converter -> Stepper Driver?
B
You do not have the required permissions to view the files attached to this post.
Re: Bryan's Fairchild 539A lathe project
Bryan you are doing excellent work with the Fairchild lathes!!
I’ve pulled mine out of storage!
I’ve pulled mine out of storage!
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Thanks for the note Kris. The whole thread from Jesus ( jesusfwrl ) was what inspired me to take on this project. If you haven't looked it over recently - it's totally worth reading again. Awesome info! Especially everything about the gear box!Kris D wrote:Bryan you are doing excellent work with the Fairchild lathes!!
I’ve pulled mine out of storage!
Fairchild/Magnetovolt Custom Disk Cutting Lathe
A quick update for tonight.
Super Glue worked! The coupler is as good as new.
I'm still fighting with slight vertical motion and will need another full weekend to chase it down. But here is a photo that shows why I wanted to mount that crazy V slot to the carriage in the back:
It's hard to see from this angle, but the idea is that I can attach anything I need to the carriage and quickly try some ideas. But if things work out well, I can lock things down.
In this shot you see I temporally eliminated the Fairchild spring from the back of the carriage and moved something equivalent to a presto style spring up front. You also see the oil dampener attached. Tonight I tried both 25 and 100 viscosity silicon dampening fluid in the dampener, and the fluids can be mixed for for anything in between. There is a calculator on line here for easy mix calculations:
http://scriptasylum.com/rc_speed/oil_mixer.html
Anyway, even with the new coupler and with WAY over-dampening at the head with 100 viscosity fluid in the dampener, I still had a slight pattern on the cut.
Where do I go from here?
This weekend I think it's worth disconnecting the motor, putting a dial indicator on the platter and looking for a pattern of vertical motion by manually rotating the input shaft to the transmission. If a pattern is seen - it's a game changer. I may just abandon suspending the head and go with a advance ball.
I looked at a few other things tonight that I'll hold off discussing until I eliminate the transmission and after my dampeners for the motor arrive...
B
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Hey guys,
Here is a little update...
I mentioned I was going to try adding dampeners to the motor suspension. Although 2 of the shocks have not arrived yet, I made progress with the other two. Here is a shot of one disassembled and one complete.
The spring and retainer are not needed, and the end cap and inner seals were discarded. The reason is that the seals cause "Stiction". Also, I milled that end off as you will see below. I'm actually running these upside down and so the top will be used for filling them.
Now, the inside plunger is shown here. This has a disc with 2 slots on the edge for fluid to flow. The big problem is that the disc is held on the rod with 2 "E" clips and the disc can move slightly between them since it's not a tight fit. I ended up gluing everything together with silicon RTV making sure to keep the slots on the side of the disc clear.(no photo).
Here is what the dampener body's look like after milling the ends off, and mounting them to some brackets. The brackets are for fit checking before I make them in aluminum, but I printed them in a material that has iron filings in it so the brackets are surprisingly stiff and dense.
You also see the rod end is attached to a round stand off. There is a reason for that! And now comes that crazy looking clamp. It's actually intended for connecting a ground wire to a pipe - but is going to do the trick on this project.
Here is a shot that shows one of the 4 springs used on the stock suspension.
And here is one of the dampeners being fit checked! I'm currently printing new bottom brackets that will have a slot so I can adjust the position a bit more, and the RTV still needs to finish curing before I fill the dashpots with dampening fluid - Hopefully later today.
That's it for now.
B
Here is a little update...
I mentioned I was going to try adding dampeners to the motor suspension. Although 2 of the shocks have not arrived yet, I made progress with the other two. Here is a shot of one disassembled and one complete.
The spring and retainer are not needed, and the end cap and inner seals were discarded. The reason is that the seals cause "Stiction". Also, I milled that end off as you will see below. I'm actually running these upside down and so the top will be used for filling them.
Now, the inside plunger is shown here. This has a disc with 2 slots on the edge for fluid to flow. The big problem is that the disc is held on the rod with 2 "E" clips and the disc can move slightly between them since it's not a tight fit. I ended up gluing everything together with silicon RTV making sure to keep the slots on the side of the disc clear.(no photo).
Here is what the dampener body's look like after milling the ends off, and mounting them to some brackets. The brackets are for fit checking before I make them in aluminum, but I printed them in a material that has iron filings in it so the brackets are surprisingly stiff and dense.
You also see the rod end is attached to a round stand off. There is a reason for that! And now comes that crazy looking clamp. It's actually intended for connecting a ground wire to a pipe - but is going to do the trick on this project.
Here is a shot that shows one of the 4 springs used on the stock suspension.
And here is one of the dampeners being fit checked! I'm currently printing new bottom brackets that will have a slot so I can adjust the position a bit more, and the RTV still needs to finish curing before I fill the dashpots with dampening fluid - Hopefully later today.
That's it for now.
B
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Long time no post, but lots of progress.
I realized that you guys haven't seen how the actual ramping function works, and decided to do a quick video that also shows the inside of the control box. Here it is:
As far as where I left off the last time I posted several weeks ago:
I ended up making new brackets and attaching the dampeners. They helped, but didn't solve the problem.
The main motor mount still vibrated on each side, but vibration in the center had been fixed by rebuilding the coupler. Then it occurred to me that the the motor run capacitor might be the issue. Since day one, the motor had a tough time running smoothly for the first 30 seconds or so after power it up. So I tested the capacitor and although the reading was correct, I tried another one of the correct value and had the same results.
Below is a video that shows what the motor was doing for the first 30 seconds or so. The clicking noise is actually the capacitor bumping against the bracket...
I decided to pull the Bodine motor, and replace it with a NEMA 23 Integrated Brushles Servo Motor.
I used a NEMA 23 right angle bracket, some rubber isolation mounts, some standard sized stand offs and washers as spacers, and made a bunch of 0.1mm shims. It took lots of trial and error to get it mounted on the existing motor bracket and shimmed to the correct height so the transmission input and motor shafts lined up.
The next trick was to make a new coupler. The transmission is 3/8", and the motor shaft is 8mm. The coupler I bought had a hard rubber spider that went between the two pieces, and it also needed to be longer. I had a machinist make me an alignment shaft so I could glue latex over the two pieces like I did when rebuilding the original coupler.
This time I got smart, and used a syringe to inject super glue as needed! Why didn't I think of that LAST time?
The servo I'm using is a little different than I have used in the past. This one (made by Leadshine) has an integrated controller with it's own DSP that deals with the closed loop. A 1,000 line (4,000 with quadrature) optical encoder is built in as well as the hall sensors. You interface to the servo with step and direction inputs like a step motor drive, and can either program the steps/rev in software (via the RS232 link) of with dip switches on the back. But it's a true servo, not a stepper - it just interfaces to the outside world like a stepper.
I decided to use an AD9833 Programmable Waveform Generator to provide the step frequency to the servo. The AD9833 is programmed via an SPI connection to the Arduino I'm using and will be mounted inside the control box as soon as the connectors arrive.
The great thing about this set up is that I can now run at 45rpm, 33 rpm and also at half speed for both! Remember, the Fairchild has a 54:1 gearbox, so to run at 33prm the motor turns at 1800 rpm. For 45rpm, the motor runs at 2430rpm.
That's the update for now.
Today I need to make cables and start on screens for speed selection and stuff. But having played with the servo and the waveform generator for a while now I'm really happy with how vibration and noise free this servo is! I think this will be the ticket.
Bryan
I realized that you guys haven't seen how the actual ramping function works, and decided to do a quick video that also shows the inside of the control box. Here it is:
As far as where I left off the last time I posted several weeks ago:
I ended up making new brackets and attaching the dampeners. They helped, but didn't solve the problem.
The main motor mount still vibrated on each side, but vibration in the center had been fixed by rebuilding the coupler. Then it occurred to me that the the motor run capacitor might be the issue. Since day one, the motor had a tough time running smoothly for the first 30 seconds or so after power it up. So I tested the capacitor and although the reading was correct, I tried another one of the correct value and had the same results.
Below is a video that shows what the motor was doing for the first 30 seconds or so. The clicking noise is actually the capacitor bumping against the bracket...
I decided to pull the Bodine motor, and replace it with a NEMA 23 Integrated Brushles Servo Motor.
I used a NEMA 23 right angle bracket, some rubber isolation mounts, some standard sized stand offs and washers as spacers, and made a bunch of 0.1mm shims. It took lots of trial and error to get it mounted on the existing motor bracket and shimmed to the correct height so the transmission input and motor shafts lined up.
The next trick was to make a new coupler. The transmission is 3/8", and the motor shaft is 8mm. The coupler I bought had a hard rubber spider that went between the two pieces, and it also needed to be longer. I had a machinist make me an alignment shaft so I could glue latex over the two pieces like I did when rebuilding the original coupler.
This time I got smart, and used a syringe to inject super glue as needed! Why didn't I think of that LAST time?
The servo I'm using is a little different than I have used in the past. This one (made by Leadshine) has an integrated controller with it's own DSP that deals with the closed loop. A 1,000 line (4,000 with quadrature) optical encoder is built in as well as the hall sensors. You interface to the servo with step and direction inputs like a step motor drive, and can either program the steps/rev in software (via the RS232 link) of with dip switches on the back. But it's a true servo, not a stepper - it just interfaces to the outside world like a stepper.
I decided to use an AD9833 Programmable Waveform Generator to provide the step frequency to the servo. The AD9833 is programmed via an SPI connection to the Arduino I'm using and will be mounted inside the control box as soon as the connectors arrive.
The great thing about this set up is that I can now run at 45rpm, 33 rpm and also at half speed for both! Remember, the Fairchild has a 54:1 gearbox, so to run at 33prm the motor turns at 1800 rpm. For 45rpm, the motor runs at 2430rpm.
That's the update for now.
Today I need to make cables and start on screens for speed selection and stuff. But having played with the servo and the waveform generator for a while now I'm really happy with how vibration and noise free this servo is! I think this will be the ticket.
Bryan
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
The correct 36 volt, 10A power supply arrived today, and the cables are done! I've had a chance to run the servo on the correct power supply and it totally works! Initial testing was with a 24V supply.
Now that I can run at 2430rpm needed for 45rpm at the platter, I'm ready to mount this bad boy with cables attached.
What I ended up doing now that I have enough power for the drive, was to bump up the step pulse rate by increasing the steps/rev dip switch settings to something well above audio but below the max step rate for the drive - shown below in the highlighted column I am using. My initial testing was at 1,000 steps/rev. Although 5000 steps/rev would work with the correct supply based on the max step rate, I decided to use 4,000 now because I know the pulse rate from the internal optical encoder is 4,000 pulses/rev. I just figure that the DSP that closes the loop would have a slightly lower error if I could keep ratios identical. It really doesn't matter me thinks. The step rate is higher than audible and well below the maximum pulse rate expected by the servo.
This servo drive can run up to 250khz on the step input (up to 500khz with some extra S/W settings). So I decided to do a little spread sheet and select a step / rev where the required frequency was below 250khz. 4000 steps/rev worked out well. Tested. Works. Done.
That's what I have for tonight...
B
Now that I can run at 2430rpm needed for 45rpm at the platter, I'm ready to mount this bad boy with cables attached.
What I ended up doing now that I have enough power for the drive, was to bump up the step pulse rate by increasing the steps/rev dip switch settings to something well above audio but below the max step rate for the drive - shown below in the highlighted column I am using. My initial testing was at 1,000 steps/rev. Although 5000 steps/rev would work with the correct supply based on the max step rate, I decided to use 4,000 now because I know the pulse rate from the internal optical encoder is 4,000 pulses/rev. I just figure that the DSP that closes the loop would have a slightly lower error if I could keep ratios identical. It really doesn't matter me thinks. The step rate is higher than audible and well below the maximum pulse rate expected by the servo.
This servo drive can run up to 250khz on the step input (up to 500khz with some extra S/W settings). So I decided to do a little spread sheet and select a step / rev where the required frequency was below 250khz. 4000 steps/rev worked out well. Tested. Works. Done.
That's what I have for tonight...
B
You do not have the required permissions to view the files attached to this post.
-
EpicenterBryan
- Posts: 738
- Joined: Sun Aug 10, 2014 9:01 pm
- Location: Eugene, OR USA
Re: Bryan's Fairchild 539A lathe project
Oh, and... Here is an optical encoder attached to the output shaft of the servo motor.
Here is the servo speed when the Arduino writes to the AD9833 to output 120Khz to run the servo shaft at 1800 RPM for 33.3rpm at the platter.
And now for 45rpm at the platter. 162Khz to the servo for 2430rpm on the output shaft.
Damn, I'm good...
B
Here is the servo speed when the Arduino writes to the AD9833 to output 120Khz to run the servo shaft at 1800 RPM for 33.3rpm at the platter.
And now for 45rpm at the platter. 162Khz to the servo for 2430rpm on the output shaft.
Damn, I'm good...
B
You do not have the required permissions to view the files attached to this post.
-
socialroots
- Posts: 161
- Joined: Tue Nov 11, 2014 2:33 pm
- Location: Dulan village,Taitung,Taiwan
Re: Bryan's Fairchild 539A lathe project
Yes,you are Byan,its all very EPIC!
maximum respect,
Patrick
P.s i wish you were my neighbor!
maximum respect,
Patrick
P.s i wish you were my neighbor!
Re: Bryan's Fairchild 539A lathe project
This is awesome!
-
venusloungerecords
- Posts: 21
- Joined: Tue Jul 04, 2017 5:24 pm
Re: Bryan's Fairchild 539A lathe project
Amazing! So inspiring! Respect to the mad skills