IDE Bug List: Atom v5.3.0.0 [Nov 28, 2006]

It appears like the links to the new IDES are up on the Basic Micro site. So I downloaded the latest versions.

After I installed the latest version of the Pro IDE 8.0.1.0. After I installed it on Vista, I have to allow it to run every time. It is like the program does not have the proper identificiation information marked on it. Is there also an update coming to the 8.0.0.0 manual?

I also checked that the Basic Atom downloads now has the link to the released version 5.3.1.0. It also has a new link to BasicAtom Manual, version 3.0.0.2. Does this correspond to the version 5.3.1.0 of the IDE?

Thanks
Kurt

Both manuals are still somewhat out of date. The 3.0.0.2 manual is better than the one before it but will have some incorrect information. We have a new person working on manuals right now but I don’t know how long it will take for him to get up to speed. Our previous writer has disappeared. He disappeared after getting out of hospital for cancer surgery and since we’ve been unable to reach him we suspect he may not be around anymore.

Not sure what you mean by the Vista problem you are having. Can you give me a better description of what you mean by “I have to allow it to run every time”

Each time I try to run the program, I get a sort-of security alert, that looks something like:

Note it was working fine with the versions of 8.0.1.0 that I downloaded earlier with the link that was in this forum message. It only started when I noticied that you had a link on the BasicMicro pages. Looking at Microsoft windows vista Help. You get this message when the program does not have a valid digital signature.

I am not having this program with the current version of 5.3.1.0.

My guess is you simply need to rebuild the program with the appropriate security junk to fix it.

Kurt

The AtomPro IDE wants access to the serial port. I suspect Vista is complaining about that. Vista likes to complain about just about everything. I’m sure it requires us to register with Microsoft(probably get WHQL certified) and pay them money to get this message to go away. Eventually we’ll go through the process but I am still developing on XP and will continue to do so until atleast SP1 of Vista.

Note: You can disable UAC(I’ve heard). I suspect you can probably even disable it just on a particular sesion or a particular program but I’m not familiar enough with Vista to know how(yet).

Hy,

I had the problem with AN converter pin numbers on the basicatom28, and the new BM atom IDE version fix it perfectly.
But since that I can’t use “pulsout” and the new update don’t do anythings …

I don’t understand, other peoples on the forum don’t look have more long time this problem. What can I do ?

thx

Gaëtan

A little more information would be helpful. How are you using pulsout? What are you pulsouting too? What exactly is the problem? Have you looked at the output on an Oscilloscope?

I’ll explain a little more of course :slight_smile:

So I’m using the basic atom 28 on the Mini ABB, and the “pulsout” is used to control the MD22 Dual 5A H-Bridge (RC mode).
Like : “pulsout pin,freq” freq is between 1000 and 2000 (1 and 2ms).

With the new BM atom board IDE version, if I use another AN converter than this one to the AX3 pin, the pulsout command may be wrong cause motors make incomprehensible things. ???
I haven’t check the output on a oscilloscope yet. During the problem, motors are always in hight speed and often in the opposite direction. Like like during a rotation… But it’s not always the same result for the same code ???
If I use only the AX3 pin or if I don’t use any AN converter, the program works correctly.
Another strange thinks the command to stop the speed and the rotation (1.5 ms) works “some times” , especially after a reset … it’s the only one.

An example of my test code :

"

vitesse = 150
virage = 150

low p0 'init des soties commandes moteurs a 0
low p1

'PP :

main

'Test des entrées

if in4 = 0 then

virage = (virage + 10) max 200

endif

if in6 = 0 then

virage = (virage - 10) min 100

endif

if in5 = 0 then

vitesse = (vitesse + 10) max 200
if vitesse = 200	then					
	vitesse = 150	
endif

endif

‘’’’’’’’’’’’’’’’’

adin AX3, 2, AD_RON, capteurir
adin AX1, 2, AD_RON, nivbat

debug [dec capteurir]
debug ", suivant, " ]
debug [dec nivbat]

debug 13 ]

'Controle des moteurs

pulsout 0, (vitesse * 10) ’ controle de la vitesse
pulsout 1, (virage * 10) ’ controle des virages

pause 750

goto main

end

"

explain continuation :

I’ve check the “pulsout” output on oscilloscope and i’ve see somethings like this :

http://www.laas.fr/~gseverac/img/pulsout_bug_oscilloscope.JPG

it’s look like if they are a short circuit or somethings else when the AN converters are used in the program.

More ever I’ve found that this problem don’t happen if there is a long time between the “adin” command and the “pulsout” command … ? At least 200 ms (if, for example, I had the line : “pause 200” in the code, the problem don’t appear)

This is really strange and really problematic so if you find a way to fix this thanks you very much !
Ask me If you need more explanations.

thx

Gaëtan

First I’ll need to know if you are running in debug mode. Debug mode will slow down your program to an unacceptable level for what you appear to be trying to do.

Second the analog capable pins on the Atom28 are P16-P19(also called AX0-AX3) but I want to make sure you have your circuit wired up correctly.

A diagram of your circuit would be helpful as well since I’m a little unclear from your description.

hy,

No I don’t test my programs only in debug mode. But there is the same problem if I try to run it in debug mode.

My connexions are like this :

http://www.laas.fr/~gseverac/img/circuitdiagram.JPG

But both motor command and analog information work well independently. It’s when I try to use it in the same time the problem appear.

thank you

Gaëtan

Gaeten, I’ll get back to this post sometime this weekend. Just wanted you to know I haven’t forgotten.

all write, thank you !

Ok, I’ve hooked up an Atom28 on one of our dev boards. I don’t have a Lynxmotion bot board(got lost somewhere). I also commented out parts of your program. here is the program as I ran it.

nivbat var word
capteurir var word

vitesse var byte
virage var byte

vitesse = 150 
virage = 150 

low p0 'init des soties commandes moteurs a 0 
low p1 

'PP : 
main 

'Test des entrées 
;if in4 = 0 then 
;	virage = (virage + 10) max 200 
;endif 

;if in6 = 0 then 
;	virage = (virage - 10) min 100 
;endif 

;if in5 = 0 then 
;	vitesse = (vitesse + 10) max 200 
;	if vitesse = 200 then 
;		vitesse = 150 
;	endif 
;endif 

''''''''''''''''' 

adin AX3, 2, AD_RON, capteurir 
adin AX1, 2, AD_RON, nivbat 

'Controle des moteurs 
pulsout 0, (vitesse * 10) ' controle de la vitesse 
pulsout 1, (virage * 10) ' controle des virages 

;pause 750 

goto main 

I’m using an analog scope. Also This is with the current release of the Atom IDE. Make sure that is what you are using as well.

First, In all cases I always get clean pulses(rising edge, high and falling edge). When I don’t comment out the adin AX1 command I DO get a little glitch in between the pulses on P1. What you would see as the low side of the pulse. However the glitch is only about a tenth of a volt above 0v. Just barely wiggles P1 when it’s low. This may be confusing your scope if it’s a digital scope or if it’s slow. I don’t know why this is happening on p1 when AX1(also called P17) is used with the adin command. I tried the other three A/D pins and did not see the glitching. I suspect it may be something inside the PIC16F876’s analog hardware. When it gets switched on it glitches and then when it gets switched off it does it again. Here is what the signal looks like on my analog scope

___________                   ____________
|          |                  |           |
|          |                  |           |
|          |                  |           |
|          |                  |           |
|          |                  |           |
|          |                  |           |
|          |                  |           |
|          |                  |           |
|          |_____\_______ ____|           |
                         /

thank you for your tests, I’ll check my scope.
But whatever the kind of scope the motor power card don’t work well when the program used the AN converter …
Have you an idea to find the problem ? Can it be because of a hardware failure ?

thx

Gaëtan

hy,
just a quick post to say that I’ve check the output on a analog scope and it’s exactly the same problem …
Have you can try with a Lynxmotion bot board ?

I don’t have my ABB available. It was packed up when I moved to California and I’ve not seen it since so I can’t test it to see if the ABB is contributing. On one of our dev boards it only shows a very small glitch and that glitch only happens when using the AN1 pin. All others(AN0,AN2 and AN3) all work without causing any glitches on the pulsing pin. Maybe someone else here with an ABB handy and an analog scope could take you cade and try it too. That would be very appriciated by both of us I think.

I would be happy to try and test this. Can you please post images of your wiring. A schematic would be better. I can say there is nothing odd about how the Bot Board handles the analog inputs. I strongly suspect a wiring error here…

I’m sorry for this late answer.

I don’t understand what you need to test the program. My wiring schematic is posted some post before.
If you need more details, just ask me and 'ill make an other scheme.

I just don’t know if it’s a wiring error because when I try with just one converter (with a GP2D12 or to a VL check) it works well …

Thanks you for your help

Gaëtan

Your schematic is incomplete. You illustrate a single line to mean a three wire connection. We can not tell if a mistake has been made. An image would be better…

all right I see,

The two other line are just used to the power supply.
The center lines are connected to the 5V with jumpers and the last line is built connect to the ground.
I think that this connections are good because independently all the the device work well.

I’ll try to find a cam today to take a picture but the board is on the rover and don’t think that will be very understandable.
If it’s too bad i’ll make a scheme more precise.

thks