New same board causing temps to be way off? Sanguinololu 1.3

(Regular reprap (cartesian?) printer, homemade. Marlin firmware, heated bed. Sanguinololu board, ATMega 1284.)

Hi, I built this printer in 2013 and at the time I understood a lot more than I remember, so please help.

What happened: My printer stopped at the very end of a print. It didn't home at the end like it always does. When I investigated, I found the X axis would only move away from home, but endstop seemed to be doing its job.

After investigation, it seemed like the Sanguinololu board had fried. It was version 1.3a.

What Ive tried:So I bought a new board, exact same, inputs and outputs, etc. It swapped straight out/in.

I put my ATMega 1284 into it thinking everything would be ok, but I keep getting low temp errors--I cant even test the movement because of the low temp errors. It thinks room temp is about 3-4 degrees now. So I thought the new board had a problem (it said it was new, but had a thick layer of dust on it, so I doubted its 'newness'), and the guy sent me another exact same board, but the problems still persist.

So last night, thinking maybe I had fried part of the chip instead of the board, I pulled the old Sanguinololu out of the junk pile, put the chip back into it and hooked it up to Pronterface. The temperatures were fine again in the old board.

What am I not seeing?

The thermistor in the nozzle and the bed read the same as each other and the same as a new unused thermistor I have, so it isn't the thermistor or its wiring. I doubt very much its the board itself being bad--two with the exact same problem seems unlikely. The old board reads temperatures fine, so I believe it's not the chip.

I am a teacher, and have students waiting for prints...

Is there a jumper or something I'm not noticing? A different resistor in the board--They all say Sanguinololu 1.3a...?

Please help, I have a classroom full of kids waiting!

Edited 1 time(s). Last edit at 02/13/2020 12:09PM by nickcartesian.

Don't know your specific board but there is a resistor associated with the thermistor (can't remember if its in parallel or series). On many controllers (if not all) its on the controller board. Its a small detail that is not always listed or noticed. If you have the wrong resistor (i think there are only 2 values, 10K or 4.7K ?) the board reads the wrong temperature. Might be worth checking.

Its a 4.7k pull up to 5v, I've never heard of a 10k variant of the Sanguinololu, but it did happen on the melzi... can't hurt to check



R9 and R10, right hand side of 644p/1284p, above the reset button



The other thing to check is you have 5v on the AREF pin. This is needed for accurate measurements on analog inputs such as thermistors

Edited 5 time(s). Last edit at 02/14/2020 10:22PM by Dust.

WOW! Thank you, I think you sent me the right direction. I checked them all: The original board did have the correct 4.7K ohm resistors. The replacement boards both have resistors in the 2K Ohm range. Whew!, I will work on getting the right resistors installed over the next few days. I can't thank you guys enough, really! Thank you SO much I am so relieved...

-Nick