Servopod Keeps Resetting

[cmmffhc]

Hi,

I am using a Servopod and it was running fine until about 15 minutes ago.

Now it just keeps resetting when I power it up.

What should I do now?

Help!!

Regards

[RMDumse]

Verify power. If the input power gets too low, it will reset constantly. It may not even get the whole log on message completed before it resets.

A bad autostart can cause the board to crash as soon as it powers up. Use the autostart bypass to see if it will stop reseting then SCRUB.

The kernel can be reflashed if you have the JTAG and kernel image.

Otherwise, tell us more of the symptoms or send it to us and we'll have a look at it.

[cmmffhc]

I have the s-files for servopod Isomax V0.82...is that the kernal image you are talking about?

I have scrub the servopod already but it keeps resetting still...

so u suggest i re flash the servopod?

[nmitech]

1. Tell us more about the power supply you are using, Voltage and current rating? High voltage input can increase the heat on the regulators and cause the thermal shutdown. Low current power supply is another main possibility for the voltage spike and cause the system reset.

2. Is there any other peripheral(s) taking power from the ServoPod as well when the 'Pod is resetting? If so, disconnect the peripheral(s) to isolate the power problem.

[cmmffhc]

Hi,

there are no peripherals taking power from the servopod.

I am using a 8V battery (i'm not sure of the current rating) but i have been using the same type of battery for the past 2 years without any problems...so i would not think it is any power supply problem...

i gather from your answers that resetting of the servopod usually is due to power problems... are there other possibilities?

Do u think it will help if i reflash the servopod?

thank you for all your prompt responses..

[RMDumse]

Yes, usually reseting problems are related to the power supply.

By comparison, the kernel in Flash is usually very stable.

Yes, reflash the s-files for servopod Isomax V0.82 using the JTAG and FLSH program. Let is know if that helps.

[cmmffhc]

Hi,

i got the following error message when i tried to flash:

C:\>flash_over_jtag flash807.cfg v082-7.S
DSP56F800 Flash loader. Compiled on Jul 23 2003, 15:47:57.
version Epsilon 0.7
(c) Motorola 2001 - 2002, MCSL
Partial Copyright 2000-2002, Zloba Alexander
JTAG IR path length: 4
JTAG DR path length: 1 (BYPASS)
IDCode status: 0xf
Jtag ID: 0xffffffff
Debug Request status: 0xf
Enable OnCE status: 0xf, polls left: 10
Enable OnCE status: 0xf, polls left: 9
Enable OnCE status: 0xf, polls left: 8
Enable OnCE status: 0xf, polls left: 7
Enable OnCE status: 0xf, polls left: 6
Enable OnCE status: 0xf, polls left: 5
Enable OnCE status: 0xf, polls left: 4
Enable OnCE status: 0xf, polls left: 3
Enable OnCE status: 0xf, polls left: 2
Enable OnCE status: 0xf, polls left: 1
Enable OnCE status: 0xf, polls left: 0
Target chip refused to enter Debug mode!
The target was reset, the application is running

[RMDumse]

So you are having trouble flashing. Have you ever flashed this kernel before successfully?

It might be time to send this board to us to look at. Is that a possibility?

[cmmffhc]

Hi,

It's the first time i'm trying to flash actually. So i hope i'm doing it correctly...

As for sending the board to you, it will be quite difficult as I am in Singapore and the shipment fees are quite expensive...

The problem is getting worse now and at the moment, when i turn on the servopod, it does not display the "Version 0.82" on NMITerm all the time and even if it displays, nothing I type is displayed...:(

[cmmffhc]

Hi,

I just tried to flash one of my Isopods, and it worked, so i would think the jtag cable is working...

Can i check with you which flash config file to use for servopod?

Is it FLASH807.cfg?

[nmitech]

Yes, that is correct!
ServoPod uses the FLASH807.cfg file

IsoPod, MiniPod, PlugaPod, TiniPod are used FLASH803.cfg file

[nmitech]

Use the volt meter and check the 3.3V regulator, VR2 make sure it has 5V on the input pin and ~3.3V on the output pin. If it does not most likely the regulator is bad.

[cmmffhc]

Hi, will check that first thing tomorrow morning, which is your midnight...

If the regulators are faulty, can i just solder new ones onto it?

[nmitech]

That is what i was hoping you can do it by yourself. Besure to check out the 5V regulator, VR1 as well. Input pin is expecting 5.5V or higher in order to get 5V out.

[RMDumse]

I'm sorry, I didn't realize you were so far away! (Or we were so far away from you, looking at it another way around.) That does change the economics of the problem. certainly.

I'm a little worried. We've had customers try to replace regulators before, mostly on the IsoPod(TM), where they get in trouble and destroy the internal layers of the board.

If you are going to pull a regulator, I'd recommend cutting its legs off first. then desoldering the individual legs, giving preference to saving the board, rather than trying to get the part out all at once.

Also, I'd be very sure about the JTAG. Be sure it wasn't rotated wrong from IsoPod(TM) where it did work to ServoPod(TM) where it did.

BTW, you know you need a different batch file starting FLASH to program the IsoPod(TM) as opposed to the ServoPod(TM) don't you? We have different kernels for each board, because of minor difference, either in internal registers, or external connections to LED's and RS-232 enables, etc.

[cmmffhc]

Hi,

I have just tried to flash another of my servopods and it turned out fine, so i should be flashing the servopod correctly.

Currently, I have 2 servopods and 1 isopod down. I'll be checking all their regulators and see if I can do anything about it...

For the 2 servopods which are down, 1 of them has regulators which becomes very hot when powered up while the other's regulators does not..

Thanks for your help again..

[cmmffhc]

Hi,

I checked the voltages...

The working one has an output of about 3.2V for the 3.3V regulator, i would think that is good.

The 2 faulty servopods has a output of about 2.5V..

I would be momving to change both the regulators, do you recommend changing any other parts?

Regards

[cmmffhc]

Hi,

I just removed the 3.3V regulator.

I found that there is a short circuit between the output line of the 3.3V regulator and the ground.

Hence I removed the 100uF capacitor that is between them. However the short circuit still exist.

I see that there is one more small little chip, but it is surface mounted, what device is that? Is it possible for me to replace that as well?

Regards

[nmitech]

I just removed the 3.3V regulator.

I found that there is a short circuit between the output line of the 3.3V regulator and the ground.


If it is a "visible" short between 3.3V & GND then most likely the solder bridge was created after you tried to remove the regulator itself.


Hence I removed the 100uF capacitor that is between them. However the short circuit still exist.

I see that there is one more small little chip, but it is surface mounted, what device is that? Is it possible for me to replace that as well?


A normal working ServoPod has a resistance of approximately ~130 ohm measured without power and nothing connecting to it.

Small 8 pin surface mount chip is the CAN bus Transceiver. I would not recommend to remove any surface mount chip at this time, unless we can pin point where the short is. If the circuitry is damaged internally from any of the IC such as the CPU chip (DSP56F807), or RS-232 chip (MAX3222), or Inverter Chip (AC05), and even bypass surface mount cap. All these can also created an invisible short from 3.3V to GND. If sending back to us for repair is not an option, then i would suggest to strip it down part by part. From through holes part first. Start from the regulators, then check for short or resistance from 3.3V & GND. Next remove C5 (100uF), and C1 & C2 caps, Q1, 3.3V bypass caps, then smallest surface mount chip next, and so on. Each time you remove part(s) check for resistance between power(3.3V) and ground. Check for visible short or burned mark on any traces, and chips, etc... Be verycareful when remove the components. Because the muti-layer board can be very easy to damage the internal layers, or pin pad, and trace .

[cmmffhc]

Hi,

I think I may really have to send the board back to you guys... what is the procedure for me to do that? And would there be any charges?

In the mean time, I'll try to do what u suggested...

Regards,

[nmitech]

please email your order info to nmiproduction@newmicros.com to request for a RMA # and instruction to return.

[cmmffhc]

Hi,

It looks like it's gonna be impossible for me to send the servopod back to u guys because it's gonna cost a bomb...$60 per piece per hour for servicing is gonna break my bank! :(

Time for me to go disect the board...pray for me some miracle happens and I can get the board working again...

Regards

[dfitz]

Hi,
I've made up a nice pcb daughter board to plug onto my plug-a-pod, and everything appears to be okay. ie: turns on, i can load on some code, ... however after about 5 mins the board starts to continually reset (ie: i see the ISOMAX... text on my serial terminal program).

From the posts here I would assume a power issue. However, I find it strange that it starts to reset after a few minutes though and not just straight away.

I started with connecting Vin to my unregulated 12-14V power input, then thought the regs were getting too hot quickly, so i tried connecting instead a regulated 5V to Vin (which is probably too low ie: < 5.5V). This is how it's hooked up currently, which might explain why it starts to reset...
I'm using the regulated 5V to power my other peripherals.

At the moment nothing else is connected except for the plug-a-pod board.
What's the best input power for this board? and am I correct in assuming that you just connect up Vin and GND to the board and it should just work (ie: you don't need to provide the regulated 5V and 3.3V cause that's done onboard).

thanks
Daniel

[nmitech]

Ideal VIN range is 6-9Vdc . You can connect your 5V regulated directly to the 5V pin on J1 instead of VIN pin. That will work.