Driving the PPM into a capacitor

1 post / 0 new
#1 23/10/2015 - 13:05
anorrie

Driving the PPM into a capacitor

We had a customer request to see what the XLC Power Module would look like with a 70uF capacatitive load.  68uF was the closest we could come up with, and here are the results.

First we output a 2vpp square wave on top of the normal 5v/12v outputs.  Here is the result with no load:

2v pp square wave with no load

  • The voltage square wave is nice and tight, with pretty square edges
  • Very little current is drawn

 

Next we added the 68uF capacitor to the outputs and ran the same test again:

2vpp with 68uF cap on output

  • The voltage pattern is still pretty square, though the rise and fall time is a fraction longer, as expected
  • The supply is producting almost 5 amps at peak on the 12v rail (hence the fast rise time)

 

Now we can do the same thing again, but with a square wave going all the way to ground, first with no load:

12vpp no load

  • We're not quite at full range here, the could go up to 14.4v and 6v at the top end, but this is the images I have right now!
  • As with the 2vpp no load image, everything is very clean

 

Now we add the 68uF cap back in again:

Full range, 68uF caps

  • First note the peak current: over 12 Amps on the 12V rail.  This is well above our 'constant' spec of 4 Amps, but can be achieved for short periods, giving a very rapid rise time
  • Even with full range rise, we see no overshoot, and very little roll-off
  • The pull down is similarly very clean, with a very fast sink rate
  • The sink rate slows a little at the end, as is expected with this design of power supply.  It still products 'nice' pattern though

 

STATISTICS

  • 2vpp with no load
    • <8uS rise time
    • <8uS fall time
  • 2vpp with 68uF load
    • <8uS rise time. 
    • <12uS fall time with 90% ocurring in the first 8uS
  • Full range with no load
    • <8uS rise time of 5v rail <12uS rise time on 12v rail
    • 5v fall time 128uS with 90% ocurring in the first 16uS
    • 12v fall time 372uS with 90% ocurring in the first 16uS

 

CONCLUSIONS

  • Rise and fall times are so fast that our 4uS sampling rate is barely able to see them
  • Fall times are a little slower than rise times (as expected)
  • There is virtually no overshoot or ripple visible in any of these tests

 

NOTE:

Remember that most drives are unlikely to appreciate a full range square wave on their voltage rails, especially not for any length of time!  You will amost certainly break something.  (We know this from experience)