본문 바로가기

_대회_Eureka_comp.

[로봇] 모터 고려_http://www.faulhaber.com/

사이트 : http://www.faulhaber.com/


DESIGN

DESIGN

Postby Tomas on Sun Nov 02, 2008 12:53 pm

Hallo,
I would like to make my own micromouse. I decide to use DC motors 1717/6V or 1724/6V or 2224/6V and IE512 encoder. I have any questions:

1.Which motor is better for micromouse. I am from the Czech Republic and Faulhaber motors are not reasonable for me probable. Reasonable are Maxon motors.

2.How expensive are motors Faulhaber 1717 / 1724 / 2224 ?

3.Use IE512 or IE256 quadrature encoder?

4.Where is possible to buy pinion and spur gear for wheel and motor? I have founded this photo:http://www.micromouseonline.com/blog/wp ... 2510-1.jpg The spur gear was made or buy ? How can I solve it ?

5.What are you use at the bottom side of yours mice? Any Teflon?

Thank you for your answers. Tom
Tomas
 
Posts: 5
Joined: Sun Nov 02, 2008 10:47 am

Re: DESIGN

Postby pittuck on Sun Nov 02, 2008 1:32 pm

Hi, welcome to the forums.

Tomas wrote:1.Which motor is better for micromouse. I am from the Czech Republic and Faulhaber motors are not reasonable for me probable. Reasonable are Maxon motors.


Erm, well, depends on how much power you require, how fast you want to go etc. etc. to be honest, any of them will work well, the 2224 (which I have) is a bit too big, but should still make a competative mouse.

Tomas wrote:2.How expensive are motors Faulhaber 1717 / 1724 / 2224 ?


All of them are going to be around about £100 each, 1717 is about £120 iirc. But then again maxon are the same prices is (probably £80 for similar size, but they lack the power the faulhabers have for the size). 

Tomas wrote:3.Use IE512 or IE256 quadrature encoder?


Depends how well you chassis is made, the size of your wheels etc. If you have large wheels (>30mm?) then you will probably need more counts per rotation, but if you want accuracy then 512 is the way to go.

Tomas wrote:4.Where is possible to buy pinion and spur gear for wheel and motor? I have founded this photo:http://www.micromouseonline.com/blog/wp ... 2510-1.jpg The spur gear was made or buy ? How can I solve it ?


I think pete makes his own gears, there are many gear manufacturers around. They tend to be a bit expensive if u need 'non-standard' sizes; I somehow ended up with 13tooth and 43tooth gears as they meant I could squeze the motors as close together as possible. As you are from the Czech Republic I have no idea where to start, technobots.co.uk do some...

Tomas wrote:5.What are you use at the bottom side of yours mice? Any Teflon?


You can use anything, even plasic screw covers (forget their names, but i think pete has a picture on his blog somewhere). If you use teflon then u will not have to replace them as often as plasic, but I am going to go out on a limb and say plastic pre made screw covers are cheaper than teflon alternatives :p

- Martyn
pittuck
Member
 
Posts: 51
Joined: Mon Jul 10, 2006 2:23 pm
Location: Reading, UK

Re: DESIGN

Postby Tomas on Sun Nov 02, 2008 2:55 pm

I have comparred parametters of maxon and Faulhaber motors, and Faulhaber motors are realy better. Maxon motors are very long.
Tomas
 
Posts: 5
Joined: Sun Nov 02, 2008 10:47 am

Re: DESIGN

Postby pranjalchaubey on Sun Nov 02, 2008 3:20 pm

Tomas wrote:1.Which motor is better for micromouse. I am from the Czech Republic and Faulhaber motors are not reasonable for me probable. Reasonable are Maxon motors.

2.How expensive are motors Faulhaber 1717 / 1724 / 2224 ?


You can check out the calculations here, 
http://micromouse.cannock.ac.uk/motors/ ... ations.htm
Important is the one used for acceleration.....your motors have enough RPMs for straight line speed anyway.

3.Use IE512 or IE256 quadrature encoder?

You'll gear down your motors by a reduction ratio of 4 or 5.....so for one complete revolution, your resolution from the encoder will be multiplied 4 or 5 times. 1280, in case you use a reduction ratio of 5 for IE256 encoders. I am pretty sure that your wheel diameter will not exceed 35mm. A little bit of calculation will show that you'll be getting enough accuracy from your encoders. Crux of the matter is that dont overspend for unnecessary resolution. I believe that a mouse can win a championship with an encoder resolution of +-1mm.:P :grin:
CP
pranjalchaubey
Senior member
 
Posts: 353
Joined: Wed Feb 06, 2008 11:57 am
Location: Noida, India

Re: DESIGN

Postby Tomas on Sun Nov 02, 2008 10:20 pm

I have tried fill this file: http://micromouse.cannock.ac.uk/motors/dcmotorcalcs.xls but i do not know two lines. (motor 2224)

Stall Current Amps ??????????????????????????????
Terminal Voltage Volts 6
Stall Torque Nm 0,0212
Free Run Current Amps 0,029

Gear Efficiency 80%
gear pitch mod 0,5
motor pinion teeth 12
wheel gear teeth 48

min desired wheel diameter m 0,0250
Number of Motors 2

Number of Cells 2
Cell Voltage Volts 3,6
Cell Resistance Ohms 0,125
Cell Capacity Ahr 0,300

mass kg 0,15
Design Speed m/s 4
Design Acceleration m/s/s 4
Motor Driver Voltage Drop Volts ??????????????????????????????
Processor/Sensor Current Amps 0,15

Can anybody check my parameters for motor 2224 ? thank you. 
TOM
Tomas
 
Posts: 5
Joined: Sun Nov 02, 2008 10:47 am

Re: DESIGN

Postby peteh on Mon Nov 03, 2008 12:07 am

For the 2224 6 Volt motor:

Stall Current Amps 3.05
Terminal Voltage Volts 6
Stall Torque Nm 0.0212
Free Run Current Amps 0.055
Motor Constant Nm/Amp 0.0070
Motor Resistance Ohms 1.97

In that spreadsheet, cells in Yellow are generally places where you can pit values to see what happens. so for the Motor Driver Voltage Drop for example, either estimate a value or work one out from the motor driver chip specs. If you are using FETs you may only have a value for RDSon so assume a worst case current and choose accordingly. This value is used to determine the maximum speed and the maximum speed at which you can apply full acceleration.

It is a complicated spreadsheet so take your time until you gat an understanding of what all the values do.
peteh
Site Admin
 
Posts: 443
Joined: Wed Jan 25, 2006 4:11 pm
Location: Lichfield, Staffordshire

Re: DESIGN

Postby Rich on Tue Nov 04, 2008 11:20 am

Cheers for the Spec on the faulhabers Pete, I have yet to use them in anger or do any serious calculations with their specs... :-D

I have some of the DSPIC33fj128MC804 and DSPIC33fj128MC802 (latter being available in 28 pin DIL) in my mouse bits collection. These have two quadrature decoders inbuilt and such parts maybe a suitable one chip solution for your mouse.

On another note I may use a CPLD and Timers as I have used this before, so its something off the shelf for me :). I may breadboard the 28 pin version and try that sometime...
God Bless thermionic Emission...
User avatar
Rich
Senior member
 
Posts: 205
Joined: Wed Feb 01, 2006 8:51 am
Location: West Midlands

Re: DESIGN

Postby peteh on Tue Nov 04, 2008 1:43 pm

I have just breadboarded up a dsPIC33FJ128MC804 (it need a shorter name). The peripheral re-mapping looks like it might mean that a 44 pin device has plenty of IO. Not sure I could get it down to a 28 pin device though. 

The PicKit2 means that I can use the same pins for programming and the UART I think.

I will next be setting up the graphical LCD and doing some motor/encoder tests.
peteh
Site Admin
 
Posts: 443
Joined: Wed Jan 25, 2006 4:11 pm
Location: Lichfield, Staffordshire

Re: DESIGN

Postby Rich on Tue Nov 04, 2008 3:36 pm

The main ambition with the 28 pin was for bread boarding... :grin: 

I guess it would be hard work - lost 4 pins with 2 quadrature channels per motor, at a very minnimum another 4 for motor control (direction & PWM) so that's 8 pins gone before you get to sensors :?

Quickly having a look at the datasheet it has 21 IO pins, but you lose two from having a crystal/resonator. So thats down to 19.

19 - 8 is 11 left. 

6 Analogue and 3 digital for sensors brings that down to 2 pins left, seeing as you need a start stop button thats one left... :( 

I reckon with a few extra IC's (mainly 4051 and 74138/74238) it maybe plausible... Using this approach you can get it sensors down to 4 pins (3 to select and trigger/reset)... Although to be honest the 804 with more pins would be a lot easier...

I'm interested to know how you get on, as its the potential route for my long awaited DC mouse....
God Bless thermionic Emission...
User avatar
Rich
Senior member
 
Posts: 205
Joined: Wed Feb 01, 2006 8:51 am
Location: West Midlands

Re: DESIGN

Postby pranjalchaubey on Tue Nov 04, 2008 9:07 pm

Hey Pete, are you planning your next mouse on MC804?
CP
pranjalchaubey
Senior member
 
Posts: 353
Joined: Wed Feb 06, 2008 11:57 am
Location: Noida, India

Re: DESIGN

Postby peteh on Tue Nov 04, 2008 9:38 pm

I am if I get the pin assignments sorted out. The dual encoder inputs are very attractive.
peteh
Site Admin
 
Posts: 443
Joined: Wed Jan 25, 2006 4:11 pm
Location: Lichfield, Staffordshire

Re: DESIGN

Postby pranjalchaubey on Wed Nov 05, 2008 5:59 am

I remember it was me who told you about them...??
I was planning to shift, but now I have jumped over to PIC32.....whats your take....
are dsPICs better than PIC32 for this purpose??
CP
pranjalchaubey
Senior member
 
Posts: 353
Joined: Wed Feb 06, 2008 11:57 am
Location: Noida, India

Re: DESIGN

Postby peteh on Wed Nov 05, 2008 7:48 am

I have not looked at the PIC32 closely enough to say. At the moment, the key reasons to choose the dsPIC are:
    Sufficient power and memory
    dual quadrature encoder inputs
    fast ADC
    flexible PWM
    easy to solder package
    re-mappble pinouts
    free programming toolsuite
    cheap programmer/debugger
    readily available

Out of that list the encoder inputs and flexible pin assignments are a big plus but not a deal breaker if they are missing.
peteh
Site Admin
 
Posts: 443
Joined: Wed Jan 25, 2006 4:11 pm
Location: Lichfield, Staffordshire

Re: DESIGN

Postby ecac on Wed Nov 05, 2008 7:59 am

My current mouse under design uses the 28pin dsPic33:

6 Analog inputs
2 sensor control lines
2 ICD2 programming lines
2 PWM channels
4 Quad inputs
2 MDir channels
1 I/O switch

2 Xtal lines. 

My PIC work is virtually non existant as I used 8051 in the past. The datasheet says you can use the internal PLL for clock generation. Two questions: 1. Does this mean you can then use the 2 Xtal lines for other I/O? and 2. If so, how fast can this PLL clock (i.e can we get 20Mhz operation)?

I am hoping to have the hardware all ready for november, but won't have anything running :-( . I am currently building a mini sumo with my daughter (she is 7) so hopefully will have that running. :grin: 

Cheers,

James
The greatest lesson in life is to know that even fools are right sometimes.
ecac
Member
 
Posts: 48
Joined: Tue Jul 08, 2008 8:30 am
Location: Hertfordshire

Re: DESIGN

Postby pranjalchaubey on Wed Nov 05, 2008 1:06 pm

ecac wrote:Does this mean you can then use the 2 Xtal lines for other I/O? and 2. If so, how fast can this PLL clock (i.e can we get 20Mhz operation)?


No, I dont think so...atleast this is not the case in PIC32. The PLL takes the input from the external cyrstal.....you need an external crystal.
CP
pranjalchaubey
Senior member
 
Posts: 353
Joined: Wed Feb 06, 2008 11:57 am
Location: Noida, India
Next  

Return to Electronics & Hardware

  

WHO IS ONLINE

Users browsing this forum: Google [Bot] and 0 guests

cron

'_대회_Eureka_comp.' 카테고리의 다른 글

[로봇] Motor Calculations  (0) 2010.01.11
[로봇] 축구 대회 키트  (0) 2010.01.11
[로봇] Min6_prof. kiat  (0) 2010.01.11
[로봇] 사운드 카드로 (오실로스코프 구현하기)  (0) 2010.01.10
[로봇] core_ARM 모듈  (2) 2010.01.06