Source for good prices on rotary encoders and knobs?

[madshaman]

Looking for better prices for a rotary encoder which has roughly same specs as this:

http://www.digikey.ca/scripts/DkSearch/dksus.dll?WT.z_header=search_go&lang=en&keywords=EM14A0D-C24-L064S%20&x=0&y=0&cur=USD

And on this type of knob:

http://www.digikey.ca/product-search/en?x=0&y=0&lang=en&site=ca&KeyWords=KN1751BS1%2F4+

I don't want to sacrifice quality or longevity, but these seem a tad expensive, would like to bring this down a little.

Thanks!

[madshaman]

Best prices I've found:

Encoder:  Arrow: 16.73, but no stock
Knob: Heilind: 19.64

I've bought a clock programmer from Arrow before, they seem nice and really seem to want business.
Never heard of Heilind before.  Anyone ever bought anything from them, website looks a bit sketchy?

[c4757p]

Huh? What's wrong with these? What exactly do you need 64 PPR optical for?

[Skimask]

My favorite pieces/parts for the past few years...

http://www.digikey.com/scripts/dksearch/dksus.dll?FV=fff40011%2Cfff802d5%2Cfffc006c&k=nav+sw&vendor=0&mnonly=0&newproducts=0&ptm=0&fid=0&quantity=0&PV235=57

[madshaman]

Huh? What's wrong with these? What exactly do you need 64 PPR optical for?

Nothing really :-), I might back off on the PPR count, but only after I've played around with a prototype.  I'm just building myself the bench supply I'd want if it existed, but if I like, might think of selling it if I can make the price competitive.

P.S. it's the freaking knob I'm a little more concerned about atm.

[c4757p]

How the hell are you finding all this expensive crap??

[madshaman]

How the hell are you finding all this expensive crap??

Not exactly trying to find expensive stuff.  Parametric search then data sheets.

Part of the price here is the longevity (2 million rotations).  You get what you pay for :-(, still $30+ for two parts isn't great.

[c4757p]

Why do you need 20 million rotations? You're making a power supply? Why don't you do some back-of-the-envelope calculations to see just how long it would take you to spin that thing 20 million times?

[PedroDaGr8]

I have that knob (except in silver), I got it as a free sample from TE. I didn't realize it was that expensive. I will say this, it's nice and it is HEAVY for a knob. Unlike most knobs, its a solid piece of aluminum (no plastic insert).  That is also what makes it expensive, its basically machined from a block of aluminum. By the way these are also called Alco Knobs or Alcoswitch Knobs (this is the company Tyco acquired I think).

If you can handle the same thing but with a black face and silver rim:
$8.50 shipped

[C]

At one time I worked with a very expensive Collins Radio. The main tunning knob was a weighted medal knob so that you could give it a spin. The Knob had a push Knob Switch so you could easily change what digit,  something  150,000 to 55,999,999. Really built for a lot of tuning and 24/7.

I do not think that knob was this good.

I do notice 500k to 1m is $10+ change.

C   

[madshaman]

Yeah, I'm obviously going for a really nice feel, the high pulse rate is really just to make the numbers scroll at a very high frame rate (also an aesthetics thing).

At one revolution per second it takes 23 days to eat up 2 million.  I have a ballpark in my head for these things ever since I was asked in public school how long it would take to count to one million.

Thing is, this spec jumps into a couple hundred thousand for the next lowest end encoder, that could be enough, but that aside, it's a given a unit which can withstand 2 million revolutions will likely be more mechanically sturdy.

I considered machining a single knob then maybe looking for someone to do the same thing on a mass scale, but that seems hassle-ridden and probably more expensive.

This is for a prototype anyway, I won't know what I'd be willing to back off on for a final design until the thing is build and I've used it.

[madshaman]

I have that knob (except in silver), I got it as a free sample from TE. I didn't realize it was that expensive. I will say this, it's nice and it is HEAVY for a knob. Unlike most knobs, its a solid piece of aluminum (no plastic insert).  That is also what makes it expensive, its basically machined from a block of aluminum. By the way these are also called Alco Knobs or Alcoswitch Knobs (this is the company Tyco acquired I think).

If you can handle the same thing but with a black face and silver rim:
$8.50 shipped

Thanks for the link, I really need the suicide knob thingy on the surface though (what's the real name?)

[amspire]

Thing is, this spec jumps into a couple hundred thousand for the next lowest end encoder, that could be enough, but that aside, it's a given a unit which can withstand 2 million revolutions will likely be more mechanically sturdy.

I am very interested in a cheap source of non-mechanical rotary switches too.

I can get mechanical rotary switches for well under $1, but they are just not going to last. If you need a bit of gear to last only a couple of years, they are probably OK.

Optical encoders should be available almost as cheaply - they just replace the contacts with a LED, two opto-transistors and a molded plastic disk. The costs of those parts to a manufacturer is probably well under 10 cents.  There is no reason for them to cost $25 or even $5. Why are they selling rubbish mechanical contact encoders with a life of 20,000 revolutions even though the hardware to survive millions of rotations was well known from the $5 trackball mice from the '80s?

[madshaman]

Thing is, this spec jumps into a couple hundred thousand for the next lowest end encoder, that could be enough, but that aside, it's a given a unit which can withstand 2 million revolutions will likely be more mechanically sturdy.

Optical encoders should be available almost as cheaply - they just replace the contacts with a LED, two opto-transistors and a molded plastic disk.

I hear you, the rub is that I could easily make one off versions of what I want that would exceed the specs of anything I could buy and would outlive the device, but not at all practical time wise, and ludicrous if thinking of selling a nice PSU at some point.

[amspire]

Thing is, this spec jumps into a couple hundred thousand for the next lowest end encoder, that could be enough, but that aside, it's a given a unit which can withstand 2 million revolutions will likely be more mechanically sturdy.

Optical encoders should be available almost as cheaply - they just replace the contacts with a LED, two opto-transistors and a molded plastic disk.

I hear you, the rub is that I could easily make one off versions of what I want that would exceed the specs of anything I could buy and would outlive the device, but not at all practical time wise, and ludicrous if thinking of selling a nice PSU at some point.

That is not what I meant. What I meant is that if a switch company can make a mechanical rotary encoder for probably about 10c a switch for a large manufacturer, they can make an optical one for 20c.

I cannot see any reason for prices to start on Digikey at about $6 and quickly jump over $20. It is just a switch housing with the same extremely cheap hardware that was in the trackball mice.

Is it just that manufacturers are now so price conscious that they will always go for the 10c switch instead of the 20c switch, even though it means that after 3 years, half their products will have a dud potentiometer?

I just want to be able to buy decent rotary encoders, and I am not sure there is such a thing as a decent mechanical contact rotary encoder.

[C]

There is a non switch, non optical option.
Found it on digikey, magnet glued to end of shaft detector ic. If memory serves, fair speed and high resolution

If you want someone to make you a knob, the raw stock and how much handling can make a big difference. On a lathe with bar stock, drill the shaft hole, machine The panel side shape, and finish outside. The lathe probably will NOT have an option for the set screw. So cut off knob. Now if you can live with the cutoff's finish witch is good, a round tip mill in a milling machine kills the center shaft left from the cutoff and may leave a small dent in center, Same mill moves out an creates finger dent. You now have a set screw less knob. That set screw will probably cost as much as the face or more to create. The trick if one, is the Need to start with a standard bar stock size and remove very little on the outside and panel side. 

C
     

[madshaman]

There is a non switch, non optical option.
Found it on digikey, magnet glued to end of shaft detector ic. If memory serves, fair speed and high resolution

If you want someone to make you a knob, the raw stock and how much handling can make a big difference. On a lathe with bar stock, drill the shaft hole, machine The panel side shape, and finish outside. The lathe probably will NOT have an option for the set screw. So cut off knob. Now if you can live with the cutoff's finish witch is good, a round tip mill in a milling machine kills the center shaft left from the cutoff and may leave a small dent in center, Same mill moves out an creates finger dent. You now have a set screw less knob. That set screw will probably cost as much as the face or more to create. The trick if one, is the Need to start with a standard bar stock size and remove very little on the outside and panel side.

C
     

Thanks (sorry, been real busy lately so haven't been around)

Yeah, the real problem is if I felt the design was good enough to share, would want off-the-shelf parts that yield the quality and feel I'm after.  Honestly I think I'll have to prototype and get feedback etc.

[madshaman]

Update: found a much cooler approach to prototyping.

My gf and I went shopping at a thrift store and I wandered back to the used ancient electronics in hope I could at least pick up some cheap transformers from the older stuff with linear supplies.

Then I noticed and remembered some the nice cases, switches and mechanics older audio equpiment from the 70/80s has.  Then I glance at the newer stuff and saw a CD playerswith a really sweet rotary encoder.  So I bought that, an old receiver/amp and an old tape deck.  Each piece cost me $10 on average and each one had at least one component, perfectly good, that would cost me more than that.

On top of that I've got three nice metal chassis (that will see my tools for sure at some point).  Some really nice heat sinks, some random IR optics (from CD player), motors actuators etc.  A nice collection of cool switches, of course random stuff like chokes, power semiconductors (which I wouldn't feel bad about killing), various sliders/pots, cables (one isolated section to another because this gear used to be high end).  Well, all sorts of good stuff.

I've been meaning to do this for myself for a while anyway, but now I'm thinking of all the different ways I can prototype (for lack of a better term) psu/test-equipment "UI" using ultra-cheap junk (that unlike today's crap is built ruggedly and precisely).

Anyway, new perspective, can worry about off-the-shelf once I have developed designs ideas more, and re-inject money into a better part of the economy, and save myself money, and save the need to manufacture new crap just becauee I myself purchased new crap.  It's win all around!

Of course now I've made this post, every thrift store will be stripped clean of all electronics with useful guts ;-)