Who is interested in a groupbuy of Mark Kelly's DC controller PCB and related parts

[Vinyl-Addict]

Originally posted by DaveM
So, I have a bit more info about how I could make the pulleys happen. I talked to the machinist at work and we looked at the shaft we are mating to (3mm~1.175"). We would start by center drilling the hole and them drilling it out. A reamer would then be run through to achieve a tight tolerance diameter. The diameter of the puley surface would then be turned. It makes no difference to the machine if it is curved or straight. From there the parts are cut off the bar stock. That should give us a runout tolerance of less than .001.
DaveM

Hi Dave, It is indeed a generous offer you are proposing. I would not take the chance your machininst is taking regarding reaming the hole. Concentricity will only be as good as the hole that is drilled prior to reaming. Personally, I would not do it. He may get lucky on a few and obtain decent results, however I would say there will be a few that won't make the grade.
He could go ahead and make a lot of extras and following inspection, throw away the rejects. Concentricity is extremely critical and I'd go for a maximum deviation of 0.0005"(.0127mm)

Ask the machinist if he has a miniature boring bar. He could bore the hole and guarantee excellent results. The boring bar I use is for boring holes under 1/8". It works very well for this project.

[john2]

I agree a boring bar is fine. A simple drill & reamer is rudimentary engineering. However, I would bore out the pulley slightly under size (say .010”) - then follow up with a 'primary' reamer (a reamer that is undersize by, say, .001"). Reamer for finish is important for the correct fit & wear properties.
Given that the motor shaft is a 'standard' size (the hardened ground motor shafts will be within, I bet, .0002" of each other)- I believe a purchased primary reamer is a reasonable investment. Thus, with this said production method, a Interference Fit heated pulley could be pushed onto the motor shaft. This idea of a grub screw fills me with misgivings! For once one tightens the grub screw you are pushing the pulley off line. If I was going to use grub screws I would use three equally distanced screws thus a measure of concentricity could, hopefully, be achieved.

[Vinyl-Addict]

Originally posted by john2
This idea of a grub screw fills me with misgivings! For once one tightens the grub screw you are pushing the pulley off line. If I was going to use grub screws I would use three equally distanced screws thus a measure of concentricity could, hopefully, be achieved.

I've been using grub screws on all of my pulleys. They work very well as long as the user doesn't get carried away by overtightening the screws.
Opposing screws 180 degrees apart work best. You barely snug down 1 screw, rotate the shaft and snug the other. Rotate once more to secure the opposing screw and you are done.

This to me is better than putting any induced force on the motor shaft by installing a press fit pulley. You also have the challenge of removing said pulley if for any reason it needs to be changed.
I realize by heating up the pulley and freezing the motor shaft, a relatively close fit can be achieved but I see more chance of harm done on the motor from this operation, obviously YMMV.

[Bobken]

Hi,

In case it helps to resolve this issue, as I mentioned in another thread a while ago, weak versions of Loctite (or similar thread-locking materials) work very well here.

This does not pull the pulley off-centre as the adhesive sets (unless there is way too excessive tolerance) and if you use 'screw' grade Loctite (the weakest) you will probably still be able to remove the pulley with a careful pull by hand, later on. If it is very 'tight', and you fear damage to the motor's bearings or shaft, a little heat on the pulley will destroy the adhesion of these weaker versions of Loctite. Not much over 100 Deg. C is needed here, so a dunk in boiling water will probably do.

'Nut' grade Loctite is slightly stronger, but I have still used it with success, even when later needing to dissemble these parts.
The heavier 'Engineering' grades are not such a good idea, as too much heat will be needed, which will likely cause some damage to the parts, if they are later needed to be separated

If using this method of pulley attachment, it is best to smear the adhesive on the *inside bore of the pulley only*, with say a wooden cocktail stick. Then assemble the parts upside down (i.e. motor on top, and pulley below), whereupon the shaft will exclude any excess adhesive out of the pulley's bore downwards, and it can be wiped off with a tissue.
If you assemble the normal way up (i.e. as used in service, with the pulley on top) and/or apply the adhesive to the shaft, as you settle the pulley down the shaft, the excess adhesive is scraped off down the shaft towards the motor's top bearing, which is not a good idea!
Maybe it is as well to ensure that the motor's top bearing is also well oiled beforehand, too, and let it 'set' upside down so any drips which could subsequently form would not enter the motor bearing.

Provided the machining tolerances are reasonable, very little Loctite is needed to secure the pulley to cope with the forces necessary during use, anyway, and I have found it will turn out to be concentric as the adhesive tries to form an even annulus between the two parts.

This saves the requirement to drill and tap the pulley for grub screws, and any possibility of 'off-centre' results, or pulley run-out, due to unequal screw tightening.

I hope this helps.

Regards,

[DaveM]

I just looked up a boring bar. We don't have one, but they are only $20 through MSC. This won't make any real difference to how we make the things. Bore...Ream... all the same to a CNC.

Now about doing a press fit/interfearance fit using heat/cooling to assemble. That is definately the best way, but I am concerned about how many of these we might send out that don't get on right. The shaft on the Maxon motors has a diameter spec of 3mm -.005 to -.01 For those of us still stuck in the dark ages of the English measurement system, that is .1181" -.0002 to .0004. The real issue with making something that small and that tight a tolerance is how to measure it.

The other question is how much expansion do you really think we will see on something that small? We could all simply use some Loc-tite bearing lock on the shafts and slide them on. No set screw to mess up the alignment.

We do have a CMM, but it is a bit nuts to try and hold 3mm -.01 spot on using a boring bar with all the wear on one cutting point. I think it would suck if someone paid for a good pulley and got one that won't fit because we were trying to hold such a tight tolerance that with a CH of tool wear the parts are no longer any good.

I guess what I am saying is... John2 you are right that this is the best way to do it, but how are we going to do it? I don't have the time to inspect each one and verify concentricity and bore diameter.

More Later... I've gotta go.

Dave

[Vinyl-Addict]

Originally posted by DaveM

We do have a CMM, but it is a bit nuts to try and hold 3mm -.01 spot on using a boring bar with all the wear on one cutting point.
Dave

Hi Dave, You can use "Deltronic pins" to check the bore diameter.
Most machine shops know what these are. There are 25 pins in a set that consist of the "nominal size" which in this case is 3mm and offer 12 pins on either side of 3mm in steps of 0.0001"

BTW, boring a 3mm hole is not that difficult, especially on a good CNC lathe, just follow basic machining fundamentals when setting up the boring bar. I make all of my pulleys in this manner.
Tool wear is of no consequence unless you use Inconel or some other highly abrasive material. I believe the pulleys were being made from free machining brass or aluminum anyway.
Also, set screws do not "mess up" alignment if the fit is performed precisely. That is what machining is all about, making precision fits on mating parts.

[GaryB]

Originally posted by Mark Kelly
I have discovered a small problem in the circuit.

A simple cure is to bypass the motor with a resistor which increases the current draw and pulls the current mirror into its linear region. A resistor of 1 k across the motor terminals improves things enormously.

Mark,
How much current is needed to stabilize things? I'm thinking I might just want to add a current source to load things instead of a resistor. Would that have any advantage over a resistor as the load?
Thanks,
---Gary

[Mark Kelly]

Originally posted by GaryB

Mark,
How much current is needed to stabilize things? I'm thinking I might just want to add a current source to load things instead of a resistor. Would that have any advantage over a resistor as the load?
Thanks,
---Gary

About 10mA total including motor current is plenty. The motor itself draw about 1 uA per rpm when completely unloaded, more when driving a platter.

There would appear to be a slight advantage for a current source in that it would draw the same current at both speed settings so the current drawn would be slightly less at the higher speed setting.

The disadvantage is that current sources are normally noisier than resistors.

The current compensation circuit can be tuned for either a resistor bypass or a current source bypass.

[DaveM]

We don't have a pin set that fine. We do have pins in .001" increments. There is nothing we do that requires that level of precision. I don't want to be responsible for creating interfearance fit pulleys. I agree with you Vinyl addict that we won't have any issue holding the diameter tight enough that the set screw would be a non issue. I do kind of like the idea of using thread lock to hold them on though. It is less work on a manufacturing end and will work fine with the amount of torque that will be needed. People can add a set screw if they would like, but I am going to use threadlock.

So is there a pulley design that people are in agreement on? Mylar or belt? Diameter? Once these things are decided, we can make this happen.

DaveM

[JRags]

Mark,

You offered to explain how to re-configure the controller for different voltages; i.e. to accommodate different motors.

if you want to know how to run the controller at higher voltage just ask.

Is there is an easy adjustment, should we want to run a 36 volt or 24 volt motor? I don't plan to make any changes in the short-term, but I would like a record of how to do it for future mods.

Thanks for all your help!