40X1 poorly calibrated in factory?

[hax]

I received 12 units of 40X1 directly from Tech Supplies in December.

I am using them to receive signals from some 08M chips, via Xbee.

The 08M chips are set to 4800 baud.

Of the 12 chips received, I found that 8 of the chips would not take a download at all.

The 4 chips that would take a download, would work OK otherwise but would not receive the serial strings from the 08M and xbee.

After a while, I found I had to use calibfreq -6 to get them to receive the serial string.

Fortunately I had 3 units of the old 40X chips and they worked no problems, which allowed me to track down the problem rather quickly.

So did I get a bad batch of chips? That is my only conclusion, but I would like some input from Technical whether there is a bad batch floating around.

I used 4mhz crystals with 22p caps after my initial internal resonator woes, but this did not fix the problem. This points to a firmware issue. As serial reception should be rock solid with an external crystal, and it rules out the 08M chips being "over clocked" because the old 40X chips worked just fine.

Having to calibfreq -6 is quite extreme. I bet the ones that don't download are too far out of the operating range to pick up the download string.

Firmware is V1.2, and I am using version 5.1.5 of the programming editor.

 

[Technical]

The 40X1 PICs internal resonators are calibrated by Microchip, not Revolution. However each PICAXE chip is tested for a download prior to despatch.

Therefore if you cannot download the PICAXE program into a brand new 40X1 it is not a calibration issue.

In this situation, where 40X downloaded but 40X1 do not, it is far more likely to be a power supply issue - the chips do use different technologies and have different voltage thresholds for signals - are you using a 3.3V supply with the XBee/PICAXE - if so this is likely to be an issue - 4.5V must be used with a 40X1 with external resonator.

 

[hax]

Hi Tech,


I'm using 5V (4.99v measured) for the 40X1, and 3.3V (3.31v measured) for the Xbee. The Xbee is connected directly to the input pin of the 40X1.

But how can the fact that setting calibfreq to -6 solves the problem, yet using an external crystal (with 22p caps) does not? This really points to a problem with factory calibration if I am not mistaken.

If only I could increase the download speed of my computer by a fraction. I bet this would wake up the 40X1 chips that are not firing up.

 

[Technical]

The download serial routines and serout/in serial routines in the PICAXE are completely unrelated in the firmware. Therefore not being able to download has nothing to do with requiring a setcalib for the serout command - the two items are completely separate.

We have had reports that others have had to use calibfreq for some serout/serin commands, and are looking into this. 40X and 40X1 do have different firmware due to the new timeout feature in serout/serin commands. But this would be a firmware, not calibration, issue. And it really has nothing to do with downloads - the two issues need to be looked at completely separately.

We did have a very small batch (under 50) of faulty 40X1s but these would have firmware reported back as 40X, not 40X1.

So do not confuse a serout/in calibration issue with a download issue, they are not the same and calibration is not an issue for the download. However if you want to send them back we can test them for you.

Can you do a firmware check - even if there was a calibration issue the firmware check would still work!

 

[hax]

Thanks for the quick responses Tech.

From memory, I tried the "firmware" button, and it timed out as well but I will do some more testing with a separate board, and separate power supply and get back to you.

 

[hax]

Just thought I would report back.

I had some time today to look into it again. I built a breadboard based minimum circuit. Only 10k resistor, 22k resistor for download cct. and another 10k resistor for the reset pin, connected to V+


I linked the two gnds and linked the two V+ and powered up via 3xAA batteries.

Tried downloading and checking firmware. Both times it times out. Also tried 4xAA batteries with two, then with one diode in series. Still nothing.

I will get another two 40X1 chips and try them. Really dont know what is going on. Remember I purchased 12 of these chips and 8 of them dont work. The other 4 work fine.