PHA / MCA setting on Cameca (vs. JEOL)... Question of the Base Line

[Julien]

Hello probers!

Over the past couple days Michael Jercinovic, John Donovan, and I were discussing over the range of values used for the SCA/MCA (resp. PHA) settings on Cameca instruments vs JEOL. To summarise and simplify the discussion and it's output, it appears that:

1) On Cameca, the bias is fixed (around 1300 V [1500 V for PCx] for low pressure counter, and around 1820 V for high pressure counter, variable...). Then, the gain is adjusted with an arbitrary value from 0 to 4095 (?) to reach a certain “recommended” value for the pulse (usually around 2 to 3 V in a scale of 0 to 5.6 V), with differences for P-10 low pressure vs high pressure (and if not, then the bias is lowered/increased to reach an adequate gain).

2) On JEOL, this is the opposite: the gain is fixed [4, 8, 16, 32, 64 or 128 - apparently higher for LDEx monochromator and for Xe-detector] and the bias is adjusted to reach an optimum pulse around 4 V in a 0-10 V range. The recommended bias should ideally be around 1700 V. If the gain increases, the optimum bias decreases.

The question is then about the setting of the base line (& window)... In each case, the operator is setting a base line (and a window if using differential). On JEOL, it is clear that the electronic noise is present when the BL is set to 0 (zero). However, according to Mike, it appears that Cameca has an hardware Baseline correction set to 560 mV, and therefore you never see the electronic noise...

[Quoting Mike J] the User Guide says that the PHA preamplifier delivers pulses from 0 to 5.6V. In integral mode, the baseline is set with a zener diode at 560mV, and window at 10V.  So the low is automatically set.  Anything you enter as baseline will be added to this 560mV, but it does not show the 560mV in the PHA display.  That makes sense as you do not see the noise at all when you run a PHA scan.  They don't mention ranges other than that specifically, but I do remember we had to put an external power supply on to get to nearly 3kV for the high pressure p10 tests (up to almost 4 bars), so I think in general you are limited to 2kV bias.

Feel free to add more details notably on the PHA / SCA / MCA setting you are using on your machine! And be aware that when you set your baseline on Cameca to 0, it is NOT really zero =)

Cheers,

Julien Allaz

[Dominik]

Hi,

interesting. I recently asked someone from JEOL whether 0-10 V means that 10 V is the maximum, like a window, or whether everything above 10 V is also registered. The answer was that everything above 10 V is also registered, but in Cameca instruments it acts like a window, i.e. the 5V essentially acts as a window.

[sem-geologist]

but in Cameca instruments it acts like a window, i.e. the 5V essentially acts as a window.

No it does not if using integral mode. In integral mode there is actually no upper limit at all, as too large gained pulses will saturate opamps at -15V or -12V (depending on generation of electronics) - in integral mode they still will be counted as pulses. Useless differential mode - is another story and there is lots of differences between SX50 SX100 (old type electronic) and SXFive. If talking about hard baseline, I don't know anything about those zener diodes (probably that is solution on SX50) as there are no such on SX100 or SXFive. On these the solution is offsetting the feedback signal to the negative side a few hundred mV with simple resistor or using offset capabilities of some integrated circuits. It is probably 560 mV with that zener diode on SX50, but on SX100 and SXFive those are for sure a bit different values (on SX100 it looks that can be adjusted, on SXFive it is fixed value). Gain goes max to 4095 as that is limits of 12bit DAC. (the maximum number of 12 bits is 4095). 5V on SXFive in PHA is max, as that is limits of the ADC. SX100 PHA ADC has limits of 10V, but for unknown reasons the signal is scaled down before being fed into that, this probably where 5.6V comes from.

[Dominik]

but in Cameca instruments it acts like a window, i.e. the 5V essentially acts as a window.

No it does not if using integral mode.

Thanks for the clarification. I was actually puzzled by that.

[sem-geologist]

Sorry for a bit off-topic.

[Quoting Mike J] the User Guide says that the PHA preamplifier delivers pulses from 0 to 5.6V. In integral mode, the baseline is set with a zener diode at 560mV, and window at 10V.  So the low is automatically set.  Anything you enter as baseline will be added to this 560mV, but it does not show the 560mV in the PHA display.  That makes sense as you do not see the noise at all when you run a PHA scan.  They don't mention ranges other than that specifically, but I do remember we had to put an external power supply on to get to nearly 3kV for the high pressure p10 tests (up to almost 4 bars), so I think in general you are limited to 2kV bias.

This (which I emphasized in bold) intrigues me a lot. I would really be interested in hearing about the results of those experiments or conclusions.