PDA

View Full Version : Chinese powerfeed question.



ScottyM
02-12-2010, 07:39 PM
My chore for tommorrow is to replace the potentiometer in my Ah-So powerfeed. Prior to it shooting craps, its working range was always down in the 0-2 area. Anything above that was too fast. I know from a long ago experience that subbing a different pot. can spread that working range out through more of it's range, but its' been long enough that I don't recall the details. I know some of you have been through this with your units. Can you help?

Thanks,
Scotty M.

whitis
02-12-2010, 08:19 PM
If it is wired as a potentiometer (3 wires used), you would need to add a fixed resistor to change the range. For example, if you have a 1K pot, replace it with approximately 200 ohm and and 800 ohm resistor to spread the lower 20% of the range over the full rotation.. If it is used as a two terminal variable resistor, just substitute 200 ohms if the speed is proportional to the resistance but you need 200ohm pot plus 800 ohm resistor if it is wired so lower resistance makes higher speed.

If wired as a pot, the total resistance across it may not matter too much, depending on the circuit. You might install, for example, a 500 ohm pot in series with a 1500 ohm resistor with a switch to short out the 1500 ohm resistor, giving you a total resistance of either 500 or 2000 ohms and two speed ranges 0-25% and 0-100%. I.E. use a pot half the original value and a fixed resistor 1.5 times the original value. If your existing pot worked fine, it wouldn't hurt to try a fixed resistor of about 3X the original value with it.

Insert the series resistor in series with the terminal nearest the fastest setting, i.e the terminal the wiper is near when turned to full speed.

ScottyM
02-12-2010, 09:19 PM
Whitis, it's time for me to reveal that I'm electronically challenged.

My pot. is indeed wired with three wires. Also there is an electronic component wired between two of the terminals. I don't know yet what this is. Could it be the resistor you mentioned? It is just some little brown dipped part. I didn't look before I left the shop to determine the resistance of my current pot.

About my only current recourse is to rely on someone at Radioshack tommorrow or if they dont have one, I have to wait until Monday to go to an electronics store that is only open during the week.

Thanks,
Scotty

RKW
02-13-2010, 12:35 AM
Radio Shack? Good luck ... "You got questions, they've got dumb looks"

Besides, you'll be luck if the even sell resistors any longer.

Just clip or desolder one leg of the unknown component and use a/your DVM to read the resistance. If the value does not quickly stabilize close to a standard value and climbs then it is most likely a capacitor, but I'm betting it is a resistor. Should not be hard to find what you need.

Black_Moons
02-13-2010, 12:49 AM
a good trick is to just buy some trimpots of a similar value and put those in series with your new pot

then adjust the trimpots as needed to fine tune the range to your exact liking :)

airsmith282
02-13-2010, 12:58 AM
that brown diped thing is called a ceramic capasitor.

resistors are color coded red green blue black and so on silver and gold bands indecate tolerance, size in diameter is wattage, the smaller it is the lesser the larger the bigger..

caps are ceramic, electrolitic,and so on,, also ratings such as pecoferrit and micoferit, pU for peco and UF for microferrit and so on

next we get into k and ohms and the list goes on,,

a capasitor stores current where as a resistor resist's current. dpending on the color bands on it determeans how much it will resist,

you might try going from a 1 k pot to a 5k or 10 k pot but in anycase you will need to know what cap you will need to use now as your resistance values are now going to change ..

Black_Moons
02-13-2010, 05:55 AM
My spelling sucks, so I should be the last one to complain but this just hurts me.

a microferrit is a very small furry animal.
http://gallery.pethobbyist.com/data/14Boo052903closeA.jpg
Id guess a pecoferrit is some kinda exotic ferret.

Microfarad (uF)
Picofarad (pF, Not pU, thats a short pun)
Note while millifarad's exist, No capacitor is ever marked in them, Capacitors marked in mFD are really in Microfarads, And no capacitor is marked in mF to prevent confusion with mFD.
Capacitor

The rest of you'r mispellings I can tollerate as my spelling sucks too, but please try and learn to spell the metric system :(

tryfred
02-13-2010, 09:26 AM
Most people go through life spelling words only one way...it is up to some of us with creative minds to offer alternatives...if they don't see the genius in that, there is little else we can do for them....

Joe

Doozer
02-13-2010, 09:50 AM
Pinkoferret

http://www.omgcritters.com/posters/easter-ferret-bunny-cute-animal-pictures.jpg

ScottyM
02-13-2010, 11:21 AM
Well, this thread certainly took an unexpected turn.

RKW, your absolutely right about Radio Shack, ... yesterday I was a bartender, today I'm your neighborhood electronics expert. However, where I live if you need an electronics part on a Saturday they are the only choice.

Black Moons, I'm afraid trimbots are something I've never even heard of. More evidence of my lack of exposure.

Airsmith, thanks for the refresher. About forty years ago I had an electronics course in H.S. and had the vaguest memories of resitors being the one with the color coded bands, but didn't trust my memory or wether things might have changed.

By the way, pinkoferret doesn't look real happy in that getup. But then again I'd have to be awfull drunk...

Thanks guys. Ill let you know how this turns out.

RKW
02-13-2010, 11:23 AM
Don't want to get into a big debate but resistors are not always marked as they should be. Sometimes there is only a value and tolerance without color bands. A meter will let you know if it isn't marked at all.


that brown diped thing is called a ceramic capasitor.

resistors are color coded red green blue black and so on silver and gold bands indecate tolerance, size in diameter is wattage, the smaller it is the lesser the larger the bigger..

ScottyM
02-13-2010, 11:49 AM
RKW, I just now got this pot out of the powerfeed. My little "unknown" is just a brown little rectangle soldered between the No.1 and no.2 legs of the pot.

I just stuck a meter on the pot and if I'm checking between 1 & 2, I only get continuity with the pot turned all the way down. When I checked between 2 & 3, I only got continuity when it was turned all the way up.

CountZero
02-13-2010, 12:09 PM
A picture would help... But the 'little brown box' sound like it might a small ceramic capacitor.

The meter is this measuring resistance or is it simply checking for continuity? What if you check between 1&3? If you are lucky the pots value might be stamped in somewhere.

ScottyM
02-13-2010, 12:21 PM
CountZero, I don't ahve a camera with me today. I just came back from Radio Duh. Just a waste of time.

I know from past dealing that I've seen these with the value on them, but this one I can't decipher. I'll list the three lines on this one.

RV24YN 20S
B103
062C

Thanks again to everybody for your efforts.
Scotty

Paul Alciatore
02-13-2010, 12:39 PM
I have two concerns to voice about this.

First, if the control is connected with three wires, it is most likely wired as a potentiometer, not as a variable resistor which would be a two wire connection (possibly with one wire connected to two terminals of the control). A potentiometer is basically a Voltage divider which has an output Voltage on the wiper (2 in my drawing) that is a fraction of the input voltage (across the other two terminals – 1 & 3). This fraction is determined by the rotational setting of the control.

http://img.photobucket.com/albums/v55/EPAIII/PotReplace.jpg

When you are replacing a control like this, it can be important to keep two different considerations in mind. First, the current available on the output pin (2) must be sufficient to drive whatever component it is connected to. If larger values of resistance are used in a replacement, the current available here will be diminished and the circuit may not work as expected. So the total resistance in the final replacement should not be much higher than the value of the original control.

Second, there is a certain amount of current that is going to flow through the control just because it is connected (terminals 1 and 3 in my drawing) across a Voltage source. This current is just wasted as it serves no real purpose in the circuit, but it is unavoidable. The Voltage source that produces this current will have limits as to the amount of current that it can provide. If a new control with lower resistance is substituted, then the current will be higher and this can cause problems in both the source of the Voltage and in the control itself as higher current means more power dissipated and more power dissipated means more heat which can slowly or even quickly destroy the control. So the total resistance of the final replacement should not be much lower than the value of the original control.

Therefore, the total resistance value of the replacement should be neither higher nor lower than the original. How do we do this?

Well, the best solution is to use a variable resistor (pot) with a lower value than the original to provide the finer control, but add a fixed resistor in series with it (on pin 1 in my drawing) to bring the final resistance back to the original value. Hence, the values of resistances A and B in my drawing should add up to the original resistance R. The ratio of A/B will give you the magnification factor for the sensitivity of the control. If A/B = 4 then the new control will be four times as sensitive but the highest speed will be only 1/4 the original.

It is usually impossible to get exact values, and in electronics there is usually a broad range of values that will work. Since we have no details of the exact circuit, it is hard to say how large this range is. I would suspect that a total resistance value (A + B) that is between perhaps 75% and 200% of the original value (R) would be just fine.

Your mystery component is probably a capacitor. I have shown it, marked with a question mark, in the most likely position in the circuit. When replacing the pot, just remove this mystery part from the old one and put it on the same terminals on the new one. If it has any markings to indicate polarity of the leads, reconnect it in the same polarity. These markings are usually things like a black band or bar by one lead or perhaps a + or – sign. If there are no such markings, it is probably OK to replace it in either orientation.

CountZero
02-13-2010, 12:39 PM
That is most likely a 10kOhm pot. I will dig a round and see if its lin or log but probably it's linear.

RKW
02-13-2010, 12:43 PM
As for the actual leads on the pot, I'm not sure they have any standard numbering. The outer leads are the full resistance of the pot the center lead is the wiper. The whole device is just a variable voltage divider. As you turn the knob, the wiper moves across the whole range allowing you to set where the tap is set.

By measuring the outer-most leads you can get the full value of the pot, that is unless it is burned out. You won't necessarily have to have the exact replacement as others have suggested. Sounds like the one you have is an indication of that since the secondary resistor was soldered in parallel. Others will work but you might have more or less variability depending on what you choose. The only thing you do have to worry about is the power rating, too low and it will smoke.

Hard to say what all the numbers mean. Could be a 20K from the first line, maybe a 10k from the second?

CountZero
02-13-2010, 12:51 PM
It is a 10k linear pot, at least according to the hits I got. Compare it to this one (http://us.100y.com.tw/US-Catalog-pdf/30-98.pdf) for example and see if it sounds reasonable.

Paul Alciatore
02-13-2010, 12:54 PM
I forgot to add how to determine the resistance value of the original control. Most pots have some kind of marking on them. It is usually in a coded manner with only two significant digits and some kind of multiplier. As an example, a 5000 Ohm pot may be marked:

5 K (the K is a 1000 multiplier)

5.0 K

5K0 (the K is both the 1000 multiplier and in place of the decimal point 5.0 X 1000)

5000

Others are possible.

Another way to determine the value is with an Ohm meter. Just measure across the outer two terminals. But be sure to do this with the device disconnected from the circuit or other components in the circuit may distort the measurement.

Paul Alciatore
02-13-2010, 01:00 PM
Another point, if it is connected as a potentiometer, as I have shown, then just changing it to another value, without adding the extra, fixed resistor, would have absolutely no effect on the speed range. The extra, fixed resistor is the actual element that provides the finer range.

Paul Alciatore
02-13-2010, 01:03 PM
RKW, I just now got this pot out of the powerfeed. My little "unknown" is just a brown little rectangle soldered between the No.1 and no.2 legs of the pot.

I just stuck a meter on the pot and if I'm checking between 1 & 2, I only get continuity with the pot turned all the way down. When I checked between 2 & 3, I only got continuity when it was turned all the way up.


You MUST disconnect it from the circuit to get a valid reading. YOU MUST.

MrSleepy
02-13-2010, 01:04 PM
Why not keep the same value of pot , just use a 10 turn one instead..

Then you wont have to worry about side effects..and you will have the extra resolution you require...
Newark Farnell stock them...bit more expensive...but cheaper than paying for repair if it went phut..

Rob

ScottyM
02-13-2010, 01:44 PM
CountZero, you just one the first prize for identification. That link id'ed it perfectly. Thanks.

Paul, I'm too unfamiliar wih this stuff. When I get home tonight I'll sit down and try to get it your instructions through my thick skull. Just to reassure you, I did indeed have it removed from the circuit. But I think this pot is dead.

MrSleepy, I'm afraid I don't know exactly what "ten turn" means. Nor do I know how to determine what my old one might have been. Please let me know.

Guys, I've been doing steel work all of my life, and fooling with machinery the last several years, but my real world experience with electronics all boils down to occasional R & R just like this. I'm always amazed at the depth of knowledge some of you demonstrate on it. Far far more than I.

Thanks again.

MrSleepy
02-13-2010, 01:54 PM
MrSleepy, I'm afraid I don't know exactly what "ten turn" means. Nor do I know how to determine what my old one might have been. Please let me know.

Thanks again.

Its a potentiometer that instead of a 300 degree arc from full value to zero has 10 turns of the shaft or 3000 degrees (ish)..

http://parts.digikey.com/1/parts-kws/10-turn-potentiometers

So you can use the same resistance on the potentiometer but with 10 times the resolution..

No Kirchoffs..microferrits or capactitaters ..

Rob

Black_Moons
02-13-2010, 02:32 PM
'103B' is likey the resistance code of '10+3 zeros' (10,000ohm, 10Kohm)
However, removing it from the circuit and using the ohm meter setting on your DMM from lead 1 to 3 is by far the best way to know exactly what your pot is.
(And yes it MUST be removed from the circuit)

Trimpot is basicly a tiny little pot, usally about 30~60 cents, requires a small screwdriver to adjust as its not really got any knob, often found right on circuit boards

http://images.google.ca/images?hl=en&source=hp&q=trimpot&oq=trimpot&um=1&ie=UTF-8&sa=N&tab=wi

http://www.solarbotics.com/assets/images/rt10k/rt10k_pl.jpg these are about the cheapest kind :)

whitis
02-13-2010, 06:47 PM
Therefore, the total resistance value of the replacement should be neither higher nor lower than the original. How do we do this?

It is usually impossible to get exact values, and in electronics there is usually a broad range of values that will work. Since we have no details of the exact circuit, it is hard to say how large this range is. I would suspect that a total resistance value (A + B) that is between perhaps 75% and 200% of the original value (R) would be just fine.
.

Good post. It is very likely that the circuit will tolerate a range of values on the pot or pot/resistor combination but it is best to stay close for the reasons you already mentioned. Which is why I suggested using a total resistance of the pot+resistor that was the same as the original value or a configuration that was either half or double the original resistance if making a two speed range model. While not as sure a bet as matching the original value, it is very likely to work. The circuit should not be too sensitive to the output impedance of the pot because a pot inherently has an output impedance that varies from 0 ohms at either extreme to half the marked value at the middle. And as for the power consumption, that isn't likely to be a problem unless they are trying to run the speed control of the motor supply which is probably around 160V.

I would suggest measuring the voltage across the two outer terminals of the pot with power applied, being careful not to touch anything (shock hazard) or short anything out with the probe leads. Wait a while after power is removed before touching anything and try to run the motor after unplugging to discharge any large caps. With the voltage reading, we can tell you what wattage you need for the new components.


And, yes, if the mystery component is connected between the middle terminal and the outer terminal that corresponds to the lowest speed setting then it is probably a capacitor.

If the original pot is 10K and 1.5watt or less, than a 5 turn pot such as this
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=SP535-10K-ND
would work if it fits. A 5 turn pot is one in which you have to rotate the knob 5 turns to go from the minimum to the maximum setting. If you expect to normally only use the lower 10% of the speed range, then you will use one turn. Down sides of a 5 turn pot is that they are more expensive, you probably won't find one locally, and you won't be able to tell by looking whether you are at, for example, 10%, 30%, 50%, 70%, or 90% of full speed without adding a turns counting dial like this (http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=H-46-6A-ND). Although it costs around $40 for a 5 turn pot and dial, that is a slick way to do it. Electrically the right part is, a drop in (well, solder in) replacement; any issues are likely to be mechanical (mounting the dial, clearance for the body of the new pot, etc)., which might be good for the electrically challenged.

Given that you apparently have a 10K ohm, pot, I would suggest these you consider these values for the several different approaches of modifying the part:
Range POT Resistor Switch
0-25% 1 turn 2.5K ohm, >=1/4W 7.5K ohm, >=1/4W None
0-25%,0-100% 1 turn 5.0K ohm >=1/2W 15K ohm, >=1/4W SPST
0-100% 5 turn 10K ohm, >=1/4W none none

Lew Hartswick
02-13-2010, 06:58 PM
And if folks wern't so paranoid about posting where they were located
there may well be an electronics tech/engineer around the corner to
give a helping hand.
...lew...

bob ward
02-13-2010, 08:30 PM
If the original pot is 10K and 1.5watt or less, than a 5 turn pot such as this
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=SP535-10K-ND
would work if it fits. A 5 turn pot is one in which you have to rotate the knob 5 turns to go from the minimum to the maximum setting. If you expect to normally only use the lower 10% of the speed range, then you will use one turn. Down sides of a 5 turn pot is that they are more expensive, you probably won't find one locally, and you won't be able to tell by looking whether you are at, for example, 10%, 30%, 50%, 70%, or 90% of full speed without adding a turns counting dial like this (http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=H-46-6A-ND). Although it costs around $40 for a 5 turn pot and dial, that is a slick way to do it.

Thanks for this Whitis, it is a very good idea.

I've always felt the speed control on my generic X feed servo is far too twitchy and coarse.

5 turns low speed to high speed and a counter will be a lot better than the present 3/4 of a turn and a graduated sticker that keeps falling off. With a fine control I'll be able to establish different speeds at different settings and make an accurate feed chart, something I have no chance of doing at the moment. Brilliant!

I've no idea yet if the components will fit in place of the originals, but if not it should be possible to turn a small housing for them and attach the housing to the servo.

I'll save $40 worth of end mills in no time.

MrSleepy
02-14-2010, 03:44 AM
Thanks for this Whitis, it is a very good idea.

I now know how Elisha Gray felt...or JS in the tailstock offset thread..

Doesnt anyone read posts above the bottom one anymore..

whitis
02-14-2010, 06:54 AM
I now know how Elisha Gray felt...or JS in the tailstock offset thread..

Doesnt anyone read posts above the bottom one anymore..

Yes, for the record, you did suggest a 10 turn pot earlier in post #22. But I think it is the combination of the multiturn pot plus dial that struck bob_ward's fancy. I almost mentioned it while writing post #2 but I was writing in terms of parts that are locally obtainable in most locations and a cost befitting a chinese import and with minimum mechanical futzing around and need for expensive dial and OP didn't seem to be concerned with keeping the upper 75% of the range. In my case, I have single turn pots but not muliturn pots with dials so one is a quick solution and the other involves waiting on parts. Even the dumb look shack has single turn pots for $3. But the idea did grow on me as I read the first 10 or 20 posts. I suggested a 5 turn to give about as much knob rotation vs speed as the OP seemed to be asking for.

10 turn pots with counting dials are great in some applications but the cost adds up pretty fast. The cost to the end user of an item will probably go up more than a $100 for each ten turn pot, so they aren't seen that often (except little trim pots) even in labs, although I have used equipment that had several on it, it cost $30K. Most of the equipment I have used with them is in the high 4 digit price tags or higher. If one is trashing end mills or parts because of speed, though, the payback time could be very short.

BTW, a 10 turn pot would turn 3600 degrees, not 3000; tends to matter in terms of turns counting dials. The dial linked to should read about 499/500 with a five turn pot and 999/000 with a ten turn pot at maximum. On a 5 turn pot and a 0-25" per minute power feed, multiplying desired inches per minute by 20 (40 for the 10 turn) may be pretty close if the speed is linear.

Personally, I am not a big fan of power feeds. Why spend $250+ for a crude power feed when I can CNC that axis for around that cost? The first is a useful accessory but a dead end and the other is a stepping stone to something bigger.

As for the absurd comparison to Elisha Gray and Alexander Graham Bell, that is another illustration of the train wreck known as the US patent system. We had two independent inventions of the laser (the first not revealed because it was classified), two for the telephone, and IIRC two for one of the major developments of the steam engine (also filed the same day) - an often cited example of an innovation suppressing patent. Yet the patent office somehow thinks trivial nuisance patents are not sufficiently obvious to someone sufficiently familiar with the art. "Method and apparatus for fine resolution speed control for machine tool power feed" would hardly surprise me, though each of us probably took less than a minute to come up with the idea when presented with the problem. Grayy and Bell's applications should have invalidated each other.

MrSleepy
02-14-2010, 07:27 AM
As for the absurd comparison to Elisha Gray and Alexander Graham Bell, that is another illustration of the train wreck known as the US patent system.


And more importantly..make sure you can trust your own lawyer..

whitis.....I wasn't having a go at you..

Rob

Paul Alciatore
02-15-2010, 01:49 PM
..... The circuit should not be too sensitive to the output impedance of the pot because a pot inherently has an output impedance that varies from 0 ohms at either extreme to half the marked value at the middle. And as for the power consumption, that isn't likely to be a problem unless they are trying to run the speed control of the motor supply which is probably around 160V.

That is not completely the case. If you simply add a series resistor to the existing pot (or a new one of the same value since the existing one is shot) then you will increase the output impedance and decrease the available current to drive the following devices. So this can be a consideration if a large change is needed in the speed range.




I would suggest measuring the voltage across the two outer terminals of the pot with power applied, being careful not to touch anything (shock hazard) or short anything out with the probe leads. Wait a while after power is removed before touching anything and try to run the motor after unplugging to discharge any large caps. With the voltage reading, we can tell you what wattage you need for the new components.

If the resistance of the original pot reads 10K Ohms or larger, just get a 1.5 Watt unit and you will be covered. If it is lower than that, then you will need to worry about the Wattage. But I doubt it as that would be a very poor design.

A multi turn pot is a good idea if you can stand the cost and if it will fit physically. The cheapiest I was able to find was about $17 and it was about three times as deep as a standart pot. A standard pot would only cost $2 or $3.

http://www.mouser.com/ProductDetail/BI-Technologies/7286R1KL25/?qs=sGAEpiMZZMvW%252b9mi%252bfqQhsFInVEqFylkOVIG1I gE2jU%3d

And if you want a turn counting knob, they are even more expensive.

http://www.mouser.com/catalog/catalogUSD/641/683.pdf

At bottom of the page. Ten turn model is about $30. With a simple knob, you would not have any indication of which turn you are on.

A single turn pot with a fixed resistor should cost less than $3, and is almost guaranteed to fit in the same space.

Paul Alciatore
02-15-2010, 01:51 PM
And if folks wern't so paranoid about posting where they were located
there may well be an electronics tech/engineer around the corner to
give a helping hand.
...lew...


Lew, I am an electronic tech/engineer. Have been for my entire career. And I'm here.

whitis
02-15-2010, 09:20 PM
That is not completely the case. If you simply add a series resistor to the existing pot (or a new one of the same value since the existing one is shot) then you will increase the output impedance and decrease the available current to drive the following devices. So this can be a consideration if a large change is needed in the speed range.


Yes, which is why I suggested either using a combination that matched the original value or one that stayed within a factor of two either way on both impedance and power consumption and never suggested just adding resistance in series with the existing value (though if the original was still good, it would be worth a try). But if the circuit cared about the impedance being relatively constant, they would have put a much higher value resistor between the wiper of the pot and the rest of the circuit. If the circuit is right on the edge to start with, then you could have a problem (which I believe I already stated) but this usually isn't the case. The likely result, though, would be a change in the non-linearity, for the better on the low speed range and for the worse on the high speed range. A result the OP probably would probably be happy with though he probably wouldn't have had the baseline data to realize he got two improvements instead of one. With the original value plus 3X or 4X series resistor, though, linearity could have taken a serious hit if it was already pretty bad.



If the resistance of the original pot reads 10K Ohms or larger, just get a 1.5 Watt unit and you will be covered. If it is lower than that, then you will need to worry about the Wattage. But I doubt it as that would be a very poor design.


The data sheet link posted earlier and the part number seems to indicate that the original was 10K ohm,1/4 Watt, linear taper. Which would indicate that the pot is running on 48V or less - or there is very good reason why it needs replacement. At 160V, it would be subjected to ten times its rated power and would have sent out smoke signals first time it was used, so we can probably rule that out. I specified the wattages for the different configurations in my table based on the maximum wattage they could be subjected to if the original pot was not subjected to wattages outside its specifications. If they didn't use 160V, it is very unlikely they used more than 30V (+/-15V). But it isn't a bad idea to check, just in case.

I think we are in agreement on the tradeoffs regarding multi-turn pot vs single turn pot.