Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Hall sensors

Status
Not open for further replies.
Hi,

How could you know wich wire is Gnd, Vcc and output. How many kinds of hall sensors are there ?

I have a DC brushless servo motor (KOKUSAN DENKI FH7-1561 TL0146) having 9 wires, and 2 suspicious ones, cuz thev're connected to the case from the outside. Then how could i test the hall sensors ?

Could there be 4 sensors instead of 3 ?

How do you test a hall sensor?

Thanks.
 
Start here **broken link removed**
 
Thanks,

I've checked before some Hall sensors datasheets, und i checked the allegro application notes, but i still can't figure it out.

Concidering it's have 3 hall sensors then we'll have 1 Gnd and 1 Vcc and 3 output wires wich gives 5, then what's the role of the 6th, unless it have 4 sensors, but what's the 4th role ???

We could admit it had 4, but how could i guess the Vcc and Gnd ? or should i reverse engeener it, I was unable to open the case then remains the bruteforce solution, I guess most of the Hall sensors starts operating at 3V then i could try powering it with 3.7V and checking the outputs, but this requires lot of tries, unless I'll be lucky and find the power wires from the first tries yet must try with the right polarity.

What would you do in a similar situation ???
 
I'd start with my Fluke Ohmmeter in the DIODE test position, measuring both directions. I could tell the power pins from the output pins.
 
I've tried with an ohmmeter at 2MΩ before, i was having responses from almost all wires. But now i tried the Diode tester here are the results:

+ probe ! - probe

black ! all except green
brown ! yellow

the green wire gave no result in both polarities but in the ohmmeter negative probe connected to the green wire gave a growing value, kind of a capacitor.

+ probe ! - probe

brown ! green orange gray ! fixed value about 1200 kind of a capacitor
yellow ! green orange gray ! fixed value about 900

Not sure, but would say:
Green ! Output A
Orange ! Output B
Gray ! Output C

and no idea about the others

what's your idea about the wiring
 
Last edited:
I opend it, almost damaged :p (damaged a bearing maybe 2 and a bit at the coupling point of the shaft) but the good news is that i solved the problem.

The wires are connected as follows:

KOKUSAN DENKI FH7-1561 TL0146

Red Phase A
White Phase B
Blue Phase C

Green Hall A
Gray Hall B
Orange Hall C

Black Gnd
Yellow Vcc

Brown I'm not sure yet, it comes from an output from a C358C amplifier (can't take the PCB away, and don't want to cause extra damage). Here begins my questions I doubt about a part marked as "04S" looks like a mini photo switch, but both sides pointing up and kind of having a resin on top instead of acrilic pointing to an encoding washer stuck to the surrounding magnet (then i'm not sure if it's a kind of hall sensor or similar, or it's a photo switch the washer is chrome then reflecting) I don't have a camera here, i would had included photos.

Then, if i'm right the 3 hall chanels outputs are enought to control it as a servo, then what is the purpose of that extra sensor ???

I hope that post will save time to someone in the future.

tschüss
 
Last edited:
I opend it, almost damaged :p (damaged a bearing maybe 2 and a bit at the coupling point of the shaft) but the good news is that i solved the problem.

The wires are connected as follows:

KOKUSAN DENKI FH7-1561 TL0146

Red Phase A
White Phase B
Blue Phase C

Green Hall A
Gray Hall B
Orange Hall C

Black Gnd
Yellow Vcc

Brown I'm not sure yet, it comes from an output from a C358C amplifier (can't take the PCB away, and don't want to cause extra damage). Here begins my questions I doubt about a part marked as "04S" looks like a mini photo switch, but both sides pointing up and kind of having a resin on top instead of acrilic pointing to an encoding washer stuck to the surrounding magnet (then i'm not sure if it's a kind of hall sensor or similar, or it's a photo switch the washer is chrome then reflecting) I don't have a camera here, i would had included photos.

Then, if i'm right the 3 hall chanels outputs are enought to control it as a servo, then what is the purpose of that extra sensor ???

I hope that post will save time to someone in the future.

tschüss

Hello mindoverflow, thanks for the info, i have one very similar to yours as label says:
DC BRUSHLESS SERVO MOTOR
FH7-1594 TLO148
(or TL0148... not sure)
D.C.V 24
KOKUSAN DENKI Co.,Ltd
MADE IN JAPAN
MFG. LOT NO. 3N04
... It has a fat 9 pin connector with the same colors as yours, being red, white and blue, bolder cables than the rest (of course for higher amperaje)
you saved me the destruction part, as they have no bolts to unscrew, so the only option is caveman method.
It's a quite heavy motor about 1 Kg.
Have you made it work?
Any commercial controller or you made it your own??
Any pic or video would be awesome...
well, thanks anyways, the most important you already helped me

regards ;)
Rodrigo
 
Hi,

I'm happy for you. Opening it is easy the cover is maybe glued or simply tight, then hitting it with the tip or a screw driver from the side of the shaft will loosen it, then remains only the shaft clip and it comes by it self (the magnet is strong enough to hold it in place).
I almost damaged the shaft because the pulley wouldn't come out then i used an angle grinder to cut it.

I thought i posted some photos, but maybe i did it elsewhere here are some recent ones.

Yes the motor works. I used my own designed circuit, all you need is a µC and a triple h-bridge. you can add a schmitt triger or comparator or what ever, but those 2 components are all you need.
You can use the microchip app note for bldc motor control (or any sensored bldc control circuit on internet) replacing their chip with the one you have anyone will do the job, you need 6 outputs and 3 inputs + what you need for your application.

Something else, the hall sensors are rated at 12v if i remember well will function at 8-9v but won't give any output at 5v. Check the datasheet.

May I ask you from where did you scavenge it, or did you buy it as is?

Good luck.
 

Attachments

  • IMG08473-20110529-1747.jpg
    IMG08473-20110529-1747.jpg
    141.1 KB · Views: 591
  • IMG08474-20110529-1747.jpg
    IMG08474-20110529-1747.jpg
    141.8 KB · Views: 592
  • IMG08475-20110529-1748.jpg
    IMG08475-20110529-1748.jpg
    106.7 KB · Views: 530
  • IMG08476-20110529-1749.jpg
    IMG08476-20110529-1749.jpg
    103.4 KB · Views: 542
  • IMG08477-20110529-1749.jpg
    IMG08477-20110529-1749.jpg
    131.2 KB · Views: 495
  • IMG08480-20110529-1750.jpg
    IMG08480-20110529-1750.jpg
    148.5 KB · Views: 570
  • IMG08481-20110529-1751.jpg
    IMG08481-20110529-1751.jpg
    111.3 KB · Views: 567
  • IMG08482-20110529-1752.jpg
    IMG08482-20110529-1752.jpg
    122.5 KB · Views: 669
hi mindoverflow,
thanx for your fast response.
Those beautifull photos gave me a nice surprise.
Did you notice the optical encoder?
Photos 4 and 5 show the encoder's "disc" with the little holes, and
Photo 2 shows the encoder's optical sensor (that you called "mini photo switch" as "04S")... at least it looks like that. So now you have encoder feedback to make easyer your motor control as a real servomotor.
It must be considered that to sense direction, encoder musts have at least two outputs (quadrature output), so maybe that sensor is double output.
Then the problem would be the need of extra cable for encoder output, and I think they those two "suspicious" cables you mentioned in the first post of this thread, but I can't see in your photos.
I was cool with the information you posted previously, but these photos tempted me to open mine to check that about the optical encoder.

I have two of these motors, they are identical. I got them as guts from an old copy machine for USD $5 (2,5 each) (hehe I was really lucky).
Yes I know a bit about the electronics stuff in fact I have a set of PIC18F2431 with QEI interface (dedicated hardware peripheral inside the microcontroller to decode quadrature encoders) and 6 advanced 14-bit PWM channels dedicated for motor control. PID tuning is my goal at this moment, but my real goal now will be to run these KOKUSAN servomotors.

I'll get a camera to send you some pics soon.

Regards, write you soon!
 
i have just checked the C358C chip you mentioned and it refers to a "UPC358C" DUAL OPERATIONAL AMPLIFIERS so it smells like outputs for a TWO channel encoder as I told you...
 
Last edited:
maybe 04s is a magnetic sensor because if it were a photoswitch then it would need a LED, which doesn't exist... if it is a magnetic sensor, these should have to be 2 of them (double), exited in some way by the metallic encoder disc holes.. well, I'm just guessing... maybe just messing, just a bit confused probably, or just right, who knows.

note: a quadrature encoder can be optical or magnetical
 
Last edited:
I admit I didn't use the 04S component output, but i believe it's a reflective photo switch rather than a magnetic sensor, and I don't believe it has quadrature output as there is only one free wire then maybe it was intended to calculate the speed.

I don't remember well, but i thing it's the reflective sensor that is connected to the amplifier only a single one is used.

And the 2 wires that i suspected at first are only a temperature sensor used to check the temp of the motor. I don't know if you have them, it's fixed with a bolt on the outside of the motor. It's a thermo-resistor, you can be sure using a multimeter to check it.

I used it maybe as a stepper rather than as a servo, but i believe it can be used as position controlled servo using only the hall sensors they could be considered as a 3 bits quadrature encoder.

I had some successful projects of pid controlling a dc servo motor using avr's better than pics I used both attiny and mega. The elm chan servo controller was of a big help(this if you don't already know about it).

May I know what model of copier did you get it, because I have a project where I need a similar one and 2 smaller ones.

ciao.
 
OK,
Being a 1 channel encoder, it allows calculate pos & vel without considering direction, and probaby that was enough as it was designed for a copy machine.
So did you finally test the remaining cable, is it really an encoder pulse output?

thx bye

PD: about the copier model I have no idea, it was already butchered at the time I got there
 
Last edited:
You can't use 1 output to calculate the position, because you can never be sure to increment or decrement the counter. Take for example if the motor oscillates between 2 steps (the motor will always oscillate when it stops) how could you know how much steps you have to increment. Specially when the encoder steps are small as in your case (compared to the coil steps).

No, I didn't check the output. I think the only reason to use it is to calculate a more accurate speed or to detect the beginning of a movement so you re-energize the coils to hold in place.

tschüss
 
Read my post #9 in this thread: I know it is needed 2 channels to output direction and so: inc or dec counter. What specifically I said in my last post (#13) was that it could be used to measure vel & pos ASSUMING DIRECTION in the specific application in the copy machine, as they know the direction in the application. Probably the specific application drives (knows) the direction and just want to control pos and/or vel.
Anyways it sounds senseless and ridiculous that a servomotor has a 1 channel encoder... unless it is designed to run in only 1 direction, or direction is known by the app.
Well also it could be used as a full featured servomotor by using coild back EMF feedback... what makes unnecesary the existence of any encoder.
Well I think it's senseless to discuss this. The fact is know what the motor is, whatever it is... :/
 
You should not assume but know what your motor is doing It should be a closed loop system, if not, it will work but will never be accurate it will be hazardous.

Anyways it sounds senseless and ridiculous that a servomotor has a 1 channel encoder...
Remember, it doesn't have 1 but 4 outputs.
 
s%#t! I know, i've just been trying to fit the reason why a servomotor has 1 encoder channel. Chapter closed.
The important thing is to test if that remaining pin has the encoder output. I will test it this night and I will tell you. ALso will try to oepn teh motor and send you some pics of that motor and other nice things i have.-
 
Last edited:
one extra hall sensor with brown color cable is for direction detection, either CW or CCW perhaps, or speed sensor as a motor feedback speed, sorry if I'm wrong.
 
one extra hall sensor with brown color cable is for direction detection, either CW or CCW perhaps, or speed sensor as a motor feedback speed, sorry if I'm wrong.
It can't be for direction detection, as there is a single output. For direction detection, you'd better check the hall sensors states, it's yet better than a quadrature encoder.

Yes it could be used as an input for monitoring speed, or detecting small position changes (as it gets triggered before the hall sensors changes state).

I simply didn't use it :)

Good luck.
 
one extra hall sensor with brown color cable is for direction detection, either CW or CCW perhaps, or speed sensor as a motor feedback speed, sorry if I'm wrong.

well... u are wrong :). To know (determine) direction, you need another led to use it as reference. Quadrature works like this:
There are two light sensors: A and B with their respective leds, where sensed light is 1 and no sensed light is 0.-
Then "if A = previousPulseB then it goes CW, if not, it goes CCW"
It means that you need movement (obviously) to "remember" B, then to compare it with A in next quadrature change.

in pseudo code:
Code:
// when any encoder change is detected ask the following
if A = B_prv then CW = 1
             else CW = 0  //you dont need a CCW variable just set CW to 1 or 0.-
B_prv = B      // backup or "remember" actual B into B_prv ("_prv" stands for previous) (actual B will be previous B in next encoder change
// direction determination done!

a more efficient code is:
Code:
CW    = (A = B_prv);
B_prv =  B;

well, that's why a single channel (single sensor) encoder cannot determine direction
good luck
 
Last edited:
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top