CryoElectronicsTesting < CryoElectronics

---+ Testing Rev 0 Cryo-Electronics Boards

2009-04-17 Board Assembly S/N 14090002, fab S/N 01

Shorted L8, so that analog/digital grounds can be connected via headers. (replace this with 0805 0ohms later)
shorted R100, so analog always on. (remove this later)
Remove C114 and C118 to fix regulator oscillations
DS2 FPGA Done is red, should be GREEN

Power

All power inputs ohmed-out, none are shorted.
applied +5.5V to digital (all analog is still disconnected, no mini-module) , draws 55mA
power good light DS4 works.
all three buck regs give correct output: 1.199V, 2.489V, 3.331V (good)
ripple of 30mVpp with 100mV spikes on top (switching)
with mini-module in place (with analog power disconnected)
- on power up draws 200mA, when FPGA done DS2 comes on draws 420mA, stabilizes to 365mA
- watchdog comes on after about one minue.
- LED1 is on on mini-module with DS2.

now connect jumpers to power analog:

analog +3.3V gives 3.288v, on immediately (new current draw is 170mA, meaning 115mA coming from analog_dig)
now short R100, so that +-4.5V come on.
-4.568V, 155mA and 4.565V, 270mA, which means +ve draws 100mA

There is a 500 Hz 1V triangle oscillation on the +4.5V and +3.3V analog power.
- this is because pin 3, bypass, on both these regs are connected via 10nF to GND. They should be connected via 10nF to VOUT. Remove C114 and C118 to fix.
-4.5V is clean

Analog drive/sense

voltage reference gives 2.502V and 1.242V (good enough, but not quite 50%)
- !!! There is a 440 kHz oscillation on the outputs, 50mVpp and 250 mVpp respectively.
- According to the datasheet, this amplifier is not unity gain stable when driving capacitive loads (we're trying to drive 400nF... this might be difficult with any op-amp). Also, it needs two volts from the rail, so it can't provide 1.25V output when V- is connected to ground. Finally it needs a supply range of 5V. I've selected opa2277 as a replacement - even this amp may not be able to drive 400nF, might have to reduce it to 4x10nF.
basic functionality of SENSE works - I inject 100mVpp at 200 Hz and see it cleanly at the output.

Clocking

Watchdog

This topic: CryoElectronics > DigitalFMux > CryoElectronicsTesting Topic revision: r4 - 2009-08-25 - GraemeSmecher