Servo Control - Atari Classics - ASM problem

[k-Pack]

I want to control 6 hobby servos but first I want to be able to control 1.

 

I've been trying to use Bob Woolley's programs as laid out in In Atari Classics January/February 1996 article. I've typed in the assembly program into MAC/65 and triple checked it for typing errors. I don't have a servo hooked up but it seems to hang up while setting up the DLI or shortly after.

 

I'm going to have to study the code a little more to understand the NOP in line 470, but what I really don't understand starts at line 530 JSR $0681, Line 580 sets the assembler's location counter to $0680, then line 620 LDA #$00. Won't this start the JSR with a BRK($00)?

 

Here's the ASM code from VOL3_2.atr that I recently found. See anything else that might help?

0100 ;SERVO.ASM    1/27/96
0110 ;BY BOB WOOLLEY  ATARI CLASSICS
0120 ;
0130 ;
0140  *=$600
0150 ;
0160 ;BASIC CALL..SET UP VBI
0170 ;
0180  PLA ;REQUIRED BY BASIC USR()
0190  LDY #$40 ;VECTORS VBI TO $640
0200  LDX #$06
0210  LDA #$07
0220  JSR $E45C
0230 ;
0240  LDA #$30 ;SETS J/S PORTS TO
0250  STA $D302 ;OUTPUT
0260  LDA #$FF
0270  STA $D300
0280  LDA #$3C
0290  STA $D302
0300  RTS
0310 ;
0320 ;
0330  *=$640
0340 ;
0350 ;VBI..WRITE TO SERVO
0360 ;
0370  LDA #$FF ;ALL PINS TO 1
0380  STA $D300
0390 ;
0400  LDY #$C0 ;DELAY FOR .5MS
0410 LPY DEY
0420  BNE LPY
0430 ;
0440  LDX #$00 ;CONSECUTIVE LOAD
0450 LP1 LDA $8000,X ;FROM $8000
0460  STA $D300 ;TO J/S PORT
0470  NOP ;INCREASES RANGE
0480  INX ;NEXT BYTE
0490  BNE LP1
0500 ;
0510  LDA $06FF ;FLAG $8000 UPDATE
0520  BNE RETN
0530  JSR $0681 ;DO UPDATE
0540  INC $06FF ;RESET FLAG
0550 RETN JMP $E462;END VBI
0560 ;
0570 ;
0580  *=$680
0590 ;
0600 ;UPDATE BIT MAP
0610 ;
0620  LDA #$00;ZERO $8000-$80FF
0630  TAX
0640 LP2 STA $8000,X
0650  DEX
0660  BNE LP2
0670 ;
0680  LDY #$07 ;STORE CONTROL BYTES
0690  LDA #$01 ;STARTING WITH BIT 0
0700  STA $06FE; BIT INDEX
0710 LP3 LDX $6C0,Y ;STOP ADDRESS
0720  JSR STORE
0730  ASL $06FE ;NEXT BIT
0740  DEY; NEW CONTROL BYTE
0750  BPL LP3
0760  RTS
0770 ;
0780 STORE
0790 LP4 LDA $7FFF,X;GET CURRENT BYTE
0800  ORA $06FE ;UPDATE THE BIT
0810  STA $7FFF,X ;STORE NEW BYTE
0820  DEX ;NEXT BYTE
0830  BNE LP4 ;UNTIL AT $8000
0840  RTS ;DONE
0850 ;
0860 ;
0870  *=$6C0
0880 ;SERVO CONTROL BYTES (ARBITRARY)
0890  .BYTE $80,$40,$20,$10
0900  .BYTE $08,$04,$02,$01
0910 ;
0920  .END

[Rybags]

I assume this example is using multiple bits which then go to a resistor ladder or DAC to create a variable voltage?

 

If this isn't what you want then you're better to start off from scratch. Also, it's not really good practice to have stuff like JMP $681 when the destination can potentially change, better to use labels.

Right you are about that JMP to a BRK. The only reason for that would be if using the Atari Asm/Ed cartridge as a breakpoint or to provide a short delay in normal operation. But using BRK to create a delay isn't good practice, normally BRK will be passed over by the IRQ handler but it's not guaranteed. I suspect it's just a programming mistake.

 

How you go about this stuff depends on how the control of the servos takes place. If it was just pure digital on demand and not going through phases or voltage ramps then you could probably just do it by directly masking and changing the PORTA bits.

[k-Pack]

I assume this example is using multiple bits which then go to a resistor ladder or DAC to create a variable voltage?

 

 

 

The program needs to output a variable width pulse. Timing is everything. I thought it would easily control up to 8 servos, and it would if it would work. The more I look at the code the less sure I am it will work. I'm switching to plain B - There is a series of 4 articles in ANTIC 12/83, 1/84, 6/84, and 7/84 by Evan Rosen. I'm hoping I can find the code I need within these articles.

 

I'll probably post the results in the "Out of the Pack" blog. Maybe before spring.

[Rybags]

Variable pulsewidth is easy enough but what would the carrier frequency need to be?

 

The thing is though, you shouldn't attempt to drive big loads from the inputs - not sure what the limits are but I'd guess in the pretty low mA range.

[k-Pack]

Variable pulsewidth is easy enough but what would the carrier frequency need to be?

 

The thing is though, you shouldn't attempt to drive big loads from the inputs - not sure what the limits are but I'd guess in the pretty low mA range.

 

The servo can be controlled with a 60hz pulse with a pulse length between 1 and 2 milliseconds. There is a circuit built into the servo that translates pulse into an angular position.

 

I will be using an external power source for the servos. I will also be looking into using an optocoupler for a bit more protection.

 

This is a link to the 4 articles that appeared in Antic magazine. So far it looks like plan B will work.

 

http://www.atarimagazines.com/index/index.php?author=Evan+Rosen&mag=antic

[+bob1200xl]

I used the ASM/ED cartridge for this program - don't know if it matters.

 

I would say that $681 is incorrect - try $680? Do you have the code that was on the disk? That should work.

 

Most of my Atari things are still packed away, but I can try to find the SERVO files.

 

Bob

[k-Pack]

I used the ASM/ED cartridge for this program - don't know if it matters.

 

I would say that $681 is incorrect - try $680? Do you have the code that was on the disk? That should work.

 

Most of my Atari things are still packed away, but I can try to find the SERVO files.

 

Bob

 

Bob,

 

I tried the $681 to $680 with no success. I'm going to try the ANTIC article and software so not sure if you need to open up the old boxes for me. Although if your looking for an excuse to open them, I'd like to see what the working code looks like.

 

I have 15 issues of Atari Classics and was thinking that was all the issues. Vol 5 -No 1 has the servo article. There doesn't seem to be any corrections in issue Vol 5 - NO 2(which is the last I own). It does mention the programs being on "Atari Classics Disk - Volume 5 & 6".

 

I got my version from VOL3_2.atr​. Now that I look at it, this disk may be from someone's collection or maybe not. The code on this disk matches the print listing.

[+bob1200xl]

I'll see what I can find next time I visit my storage lockers...

 

 

Bob