Re: IXYS High Voltage current regulator



Winfield wrote:
Winfield wrote:

One alternate is Supertex DN2540N5 depletion-mode MOSFET
in a power TO-220 package, in production, and stocked by
Mouser,http://www.supertex.com/pdf/datasheets/DN2540.pdf
They don't show a very detailed Id vs Vgs plot, but I put
some of these on the bench to get the data. One point:
unlike the sloppy JFETs of old, these modern MOSFETs seem
to have much less Vgs spread from part to part.

At low "sub-threshold" currents the dn2540's drain current
rises exponentially, just like a BJT transistor, but with
half the slope, 130mV/decade compared to 60mV. -2.75 volts
gives about 100uA. Above 1mA the rising-current plot tails
off, and we get to our 50mA goal at -2.15 volts. A 43-ohm
current-setting resistor would do the trick.

positive ------, 50mA high-voltage current source
|
d dn2540n5
,-- g
| s --/\/\--+---- negative
| 43 |
'---------------'

In taking these measurements, I was surprised to find the
'2540 really wanted to go into RF oscillation in the linear
mode, even at rather low currents like 5mA and low voltages
like 2.5 volts, so I was forced to solder a 0.1uF Vgs cap to
the part. Likewise, the self-bias circuit above oscillated
so badly above 25 volts that my DVM made faulty readings.


It's likely a better approach would be simply adding
a gate resistor.

positive ------, 50mA high-voltage current source
|
d
,-- g
| s --/\/\--+---- negative
| 100 43 |
'---/\/\--------'


We can hope the ixcp10m45s will shine in this regard ...
http://ixdev.ixys.com/DataSheet/98704.pdf


The ixcp10m45s I measured has a substantially different
Id-vs-Vgs plot than the dn2540n5, with a strong tailing-
off of the drain current (and reduced g_m). This is a
surprise, because the 10m45 apparently has a much larger
die than the 2540, as is evident from its nice low 3.1
C/W thermal resistance, vs 8.3 C/W for the dn2540. On
this basis alone the 10m45s would be preferred.

positive ------, 50mA high-voltage current source
|
d ixcp10m45s
,-- g
| s --/\/\--+---- negative
| 100 34 |
'---/\/\--------'

I measured a nice stable 50mA current into a 15-ohm load
(dropping 7.5V, similar to the string of tubes), over a
voltage-drop range of 5 to 260 volts for the ixcp10m45s.
There was no sign of oscillation with the 15-ohm load.
(As reported, there was oscillation at high voltages if
loaded only with a 1-ohm multimeter 200mA current-range.)

If one of these setups was used for a 90-volt drop, the
FET would need to dissipate 4.5 watts without junction
overheating, which would require a serious heatsink
plus good airflow. It might be wise to add a 1k-ohm
10W power resistor in series to drop half the voltage.


Thanks for taking the time to make all those measurements, Win.

.

Relevant Pages

  • Re: MOSFET answer
    ... with 2.7 volts on the gate, the answer appears to be IRF3708. ... components that's just exploding in options, ... It's also important to rely on guaranteed data and for Rdson at a certain Vgs that would be the "max" column. ...
    (sci.electronics.design)
  • A spectre newbie question
    ... I am kind of stuck as although I sort of know how to plot gm/id vs. Vgs ... I wonder if there is anything like this in Spectre? ...
    (comp.cad.cadence)
  • Re: A spectre newbie question
    ... I dont know how to sweep Vgs-Vt instead of just Vgs ... (or change the axis variable using the waveform calculator....). ... I am kind of stuck as although I sort of know how to plot gm/id vs. Vgs ...
    (comp.cad.cadence)
  • Re: A spectre newbie question
    ... I am using Spectre now in Analog Design ... I am kind of stuck as although I sort of know how to plot gm/id vs. Vgs ...
    (comp.cad.cadence)
  • equations on appropriate graph scales
    ... I need to solve the following question by excel: ... Plot the following equations on appropriate graph scales ... Variable y has units of volts ...
    (microsoft.public.excel.misc)