Is my reasoning ok?

[JavaRN]

After long hours of debugging I managed to get my SP03 working however I am still not convinced of what I have done and I need your opinion. The problem was that my speech synthesizer SP03 was continuously rebooting whenever I inserted MOTOR commands in my programs otherwise it worked fine. After thorough testing and even help from people in this forum (especially NovaOne - thanks Chris!!) I still couldn't get the blessed thing to work - it was all the time rebooting. So I decided to contact Hitec help and beleive me they were extremely helpful. They asked me to send the code I was using and they tested it on a board with I2C connection and the code was fine - no errors, everything was working perfectly. So the they suggested that I use external power for the SP03 or use the servo connector ports. So I connected the SP03 power to the servo connector ports and it worked perfectly!!!!

I did this because I felt that the SP03 was not getting enough power when all the servos where switched on (through the MOTOR command). I also noticed that even when I connected the sparkfun bluetooth module the same thing happened so I came to the conclusion that the SP03 was not getting enough power and as a result of this it was rebooting continously. That also explains why when I didn't use the motor commands it was working fine since there were no motors consuming powers.

Is my reasoning right or am I missing something?


Charles

[i-Bot]

The 5v supply normally holds up OK. What other devices do you have on the 5V rail (Gyro, bluetooth, heated gripper mits) ?

The SPO3 specifies a 5V supply voltage, have you supplied it direct with the servo power. I don't have a schematic for the SP03, but although there is 3.3V regulator for the WTS701, I think the PIC is powered direct from the 5V. Or do you have a seperate 5V LDO regulator for the SP03 ?

Maybe your battery is starting to fail

[NovaOne]

Sounds reasonable....I had to connect a 1000uf capacitor to my 5v supply, but only to remove background noise.

I would still regulate the voltage for the SP03 to 5v, maybe using a 7805 3 pin regulator?...just to be on the safe side?

I'm really pleased you got it all working in the end, sorry i couldn't have been more use.

Chris

PS I'm still waiting for my ceramic/piezo amp to arrive.

[JavaRN]

The 5v supply normally holds up OK. What other devices do you have on the 5V rail (Gyro, bluetooth, heated gripper mits) ?

I have bluetooth (Sparkfun), and four sensors - three IR and one Sonar, and also a Tilt sensor (I'm almost full with sensors!)

The SPO3 specifies a 5V supply voltage, have you supplied it direct with the servo power.

Yes.

Maybe your battery is starting to fail

Fully charged, my battery marks something like 6.5V on the multimeter.

One final question - Do you think that the SP03 might blow up if I leave everything as it is or get a new battery. (because now that I got it to work I don't want to blow it up!!!)

Charles

[i-Bot]

I think you really do need a 5V regulator. Try to get a low voltage drop out regulator.

Did you try the big capacitor suggested by NovaOne on the 5v rail ?

[JavaRN]

I tried the 7805 3-pin voltage regulator this morning. After a full charge the SP03 works for some time then back to reset, sometimes it works sometimes it doesn't. I've noticed that with a full charge the battery marks 6.9V but then drops after sometime to 6.4V (by this time the SP03 starts rebooting) then to 6.2V. Do you think the battery needs to be replaced?

Thanks

Charles

[NovaOne]

Sorry, my fault...a 7805 ideally needs 6.8v Input voltage......try a low voltage drop regulator as i-Bot suggested. Eg a LM1086 CS 5.0.(1.2v drop)....or better still one of the MIC29 range (max 0.4v drop), that should buy you a little more time.

Sorry the SP03 didn't turn out to be "Plug-Code and Play" for you.

Unless you really have a bad battery or some other fault, it sounds like you need extra batteries or a higher capacity main battery.

Wasn't it you who was looking at a 2AH battery a while back?

Chris

[JavaRN]

Sorry the SP03 didn't turn out to be "Plug-Code and Play" for you.

When I started this hobby, six months ago I knew that it was going to be a little difficult from the electronics side, but I am learning . . .

Unless you really have a bad battery or some other fault, it sounds like you need extra batteries or a higher capacity main battery.

What kind of fault? I don't mind buying another battery but if there is a fault I don't know how to deal with it!!!

Wasn't it you who was looking at a 2AH battery a while back?

Yes I was looking for a battery a while ago in fact I ordered one and I am waiting for it to come.

Thanks again all for your help I keep you informed with my progress.

[NovaOne]

What kind of fault? I don't mind buying another battery but if there is a fault I don't know how to deal with it!!!

The best way to determine if you have a "fault" that's draining the battery, is "simply" to compare the current being drawn by an identically set up but guaranteed working RN....Easy

But seriously if everything is working, I think the most common fault on my RN (and a good few others from what I read) is insulation worn off (or trapped) servo wires.
Unless you have an infra-red camera handy?

My RN power rail voltages look just like yours, this is 2mins after power up:


Chris

[JavaRN]

Wow that's interesting! Why don't you post some pictures of your RN it looks different! - I mean - the back is transparent and there might be other things worth seeing. Having said that I have a problem closing back my RN there are so many wires that I just can't close the lid back! To make things worse I had to remove the backpack because it got heavier (with the SP03) and the robot was falling backwards almost all the time! Any suggestions to replace the back part of the robot (where the wires are).

Charles

[NovaOne]

In fact the back cover was the first mod I did to my RN, because of the limited space. Its made of 2mm Polycarbonate (lexan if you prefer the brand name).

Make sure all the edges of where your bends will be are smooth (or it will crack), then bend it COLD. (Unlike acrylic, heating makes Lexan brittle)

The LCD volt meter was instead of a LED bargraph, since it is much more efficient, its an DPM ANDERS OEM-33 from RS Components Part No. 260-303, main supply current is only 300uA.

Why don't you post some pictures of your RN it looks different!

I'd rather not take any more pictures yet due to so many half started upgrades (eg cameras etc) "bits" hanging off....

The only other mods I haven't posted any details of are the 2.5mm LCD-PC socket for debugging, under the display to the right, the "hearing" mics and LDR, which are nothing new.

When I've successfully made a ceramic amp for the Sp03 I will be mounting the Sp03 PCB in the chest, along with a extra 16 port i2c i/o board (half finished) based on the MCP23016....if it works. I need the additional i/o because except for the PWM outputs (can they be used as inputs????) I have no spare ports: ie

4 analogue ports used for Gyros, 2 for IR proxi's, 2 for accelerometers
19 digital ports used for servos, 2 for i2c, 2 for sound sensors, 1 for light sensor.

Why do I need 16 more digital i/o lines.... well I want to put lots of sensors on the feet, to stop RN stepping off a table for example. Also I've been wondering if it would help if the feet had pressure sensors and were articulated (ie had a toe "flap" section ) to enable RN to walk more naturally, not Flat Footed.....

Well that last bit was all pie in the sky.....just dreaming

Well Charles thats about it for my RN.

Having said that I have a problem closing back my RN there are so many wires that I just can't close the lid back! To make things worse I had to remove the backpack because it got heavier (with the SP03) and the robot was falling backwards almost all the time! Any suggestions to replace the back part of the robot (where the wires are).

My only suggestion for you is to balance the load on RN, rather than putting it all in the backpack. My main thought with any mod is will it affect the running of the template program routines....because they take so long to sort out if it does.

Chris signing out.

[JavaRN]

Unfortunately I couldn't find a low voltage drop regulator in the electronics shop and someone suggested using the 7806 instead of the 7805. What do you think?

Charles

[NovaOne]

Unfortunately I couldn't find a low voltage drop regulator in the electronics shop and someone suggested using the 7806 instead of the 7805. What do you think?

Charles

A 7806 would not help at all, it would need about 7.5volts to function properly.

As a stop gap measure only, if you really can't get a regulator put at least one 1n4001 diode in series, with the 6.5v line, it will add a bit of protection and should reduce the voltage onto the Sp03 board.

I suggest you order a MIC29202BT 0.4A from:

http://www.rapidonline.com/

Chris

[JavaRN]

Thanks again for your feedback I will start with the 1000uf capacitor first to see if it makes any effect. By the way, excuse my ignorance, can it be because I used 4K7 resistors on the SP03 board for SCL and SDA that all this is happening?

Charles

[NovaOne]

I doubt it, a 4k7 will draw 1.7mA less than a 1k8

In fact i was considering increasing the value of my pull-pulls to 4k7

Chris

PS
Please keep us updated on your vision system progress.

With your experience, can you think of any simple processing techniques i can do on board with a microcontroller?