AXE033 Volts

[jaka]

I have built a data logger using snippits gleaned from this forum. It works ok.

Setup is 18X/axe033/24lc256/3.3v regulator/ 4 x nicads..

Although i can display the correct readings on the LCD and using debug, ie 3.3v = 255, using a digital voltmeter the supply is 4.5v.

LCD is connected to nicads. So no rocket scientists required. Pullups are coonected to 4.8v.

Possible answer:- zap 2 x 4k7's on 033. put 2 x 4k7's next to 18x, ie 3.3v.

Has anyone tried this, if so which ones are they on the 033. There are 3 x 4k7's.

Tracks difficult to follow and I would need to unsolder the serial to look underneath.

Can't find any info if I do a search.

 

[eclectic]

Jaka.

I’m not sure I understand your question, but the answers might be
found, using a conductivity test and

Page 6. Section 2)

http://www.rev-ed.co.uk/docs/AXE033.pdf

and Page 1 top right

http://www.rev-ed.co.uk/docs/DS1307.pdf

e.

 

[jaka]

Thanks for your response electric.
as you may know, the AXE033 has two 4k7 resistances on board, these are required by the serial chip/I2C bus and I am presuming that this chip has open drain inputs/outputs.

The AXE033 has a 4.8v supply (5.6v if freshly charged). The supply to the 18x is regulated 3.3v.

If I unplug the AXE033, the supply drops back to 3.3v. I figure that this is spurious signals getting back to the supply rails via the I2C bus and my volt meter is catching them (don't have a scope).

When reading a A/D signal the supply must drop back to 3.3v because I am getting accurate readings ie there must be a backfeed to the regulator via the 18x. I have 2 100nf caps each side of the regulator. All are within 10 mm of the 18x. Also this spurious voltage rises to nearly 5v with fresh batteries.

My questions are then, if I clip through the resistances and put two on the 18x board will it still work, because the I2C bus will be at 3.3v and the serial board is at 4.8v. If so which are the two resitances.

I need the layout drawing for the AXE033. Possibly only technical can help here, unless someone else has been there.

 

[eclectic]

Jaka.
Sorry I can't help with the tech. queries.
However, I've found which two 4k7 resistors
connect to SCL and SDA on my board.

I'll leave you to make your own decisions
and experiments.
e.

 

[BeanieBots]

Another approach would be to use the AXE033 in serial mode and simply not connect the I2C lines thus avoiding the problem.

 

[hippy]

I cannot see how PICAXE +V (Vdd) can be 5V but internal ADC Vref be 3.3V, so maybe rocket scientists are needed

SCL and SDA should be open-collector, but the internal circuitry is probably much more complex than the block diagrams show. There are clamping diodes between the I/O lines and Vdd.

I'd try with the 4K7's from SCL and SDA to 3.3V, their values may need adjusting, or use serial as suggested.

Simpler still; run the PICAXE from the same supply as the LCD.

 

[jaka]

Hippy: Tha reason for having the 3.3v regulator is that I need a stable reference volage. If I used a 5.0v regulator then I can't use nicads, unless I use 6 cells.

The idea is to have a compact unit that I can hide away in an enclosure, so that it would be difficult for someone to spot that I am snooping.

The way I read the spec on the AXE033 is that it is only possible to write to the RTC. I see little point in having a data logger that don't know the time. Maybe I have read it wrong.

I think I may have to remove the chips and bell out the connections. then cut through one end so that I can resolder them.

 

[jaka]

The third line in my last post refers to using serial for the AXE033. Using I2C allows me to read the time and act upon it ie store value in 24LC256.

 

[hippy]

The DS1307 needs 5V but there may be alternatives which can wire to the 18X directly at 3.3V, or even connect a 32kHz watch crystal and have the 18X run its own equally accurate RTC ( that's a bit involved but not too hard ).

It should be possible to use the NiCd supply unregulated and use another ADC to read a reference voltage, determine what the supply voltage is and the measured voltage that way.

A more detailed description of what you are "snooping" on and what you need the black box to do may turn up other solutions.

[Edited - thought the DS1307 could be 3.3V, but no ]

Edited by - hippy on 15/07/2007 13:28:35

 

[jaka]

Reason for logger: read 4-20mA loop in process control to determine what is wrong with the plant. Can be pH, flow, dissolved oxygen, pressure etc. Say every 15 mins.

Don't forget I am using the AXE033 so I need 5v for the LCD. Even putting the RTC on the main board won't cure the problem.

I have just had a thought. Put the RTC on the main board and use serial for the 033. I can then disconnect the display when not required.

But just snipped two 4k7's out. Will put two on the main board as a first approach. Try that later. Let you know of the result.

 

[jaka]

Hippy: That was a failure. Guess sda and scl is not a simple pullup open collector/drain cofiguration as you suggested, I was still getting 4.5 volts.

I am going to take your advice, add two more cells (7.2v) and make the whole thing 5v.

Probably was not to good for the AXE033 anyway. Just discovered that four fully charged nicads can get uo to 6 v.

Thanks for the assistance.

 

[BeanieBots]

Interesting result.
I've never used the AXE033 in I2C mode let alone attempted to remove the 4k7 pull-ups but I can assure you that the DS1307 is indeed open collector and CANNOT drive the sda scl line high. I can only conclude that something else on the AXE033 is pulling the lines high.

 

[jaka]

Agreed.
If I remove the 1307 then everything is fine. I could look at the data sheet for the pic they use, but I can't read it (eyes are fading in my old age) it probably would'nt tell me much anyway.

I will settle for the easy option and put it down to experience + take note of what hippy say's in furure.