Function generator in series with a power supply

[dadler]

Are there any gotchas when putting a function generator in series with a DC supply?

Any risk if the current draw is significant?

I'm working on a power supply and trying to simulate common mode noise/mains noise/etc, such as what might be seen after noisy AC rectification.

Is there a better way to go about this?

I found the following document ( http://cp.literature.agilent.com/litweb/pdf/5990-3466EN.pdf ):

If you want to produce a signal riding on top of a DC offset,
but you require a DC offset larger than your function generator can
deliver, you can put a DC power supply in series with the function
generator output, as shown in Figure 3 above. Using this technique,
you will have the full bandwidth capabilities of the function
generator available to you in addition to having fl exibility in the
DC level provided by the power supply. However, this technique
has a few important limitations. The output of the function generator
may or may not be internally isolated (fl oating) from earth
ground. When it is isolated, there is a specifi cation indicating the
maximum amount of voltage off of earth ground that you can fl oat
the output. For example, the Agilent 33220A’s output is isolated
from earth ground and you can fl oat the output off of ground up to
+/- 42 V. This means that if you put a DC power supply in series
with the function generator output, the DC offset must be less than
+/- 42 V. If the function generator output is internally connected
to earth ground, then the power supply output voltage must be
isolated from earth ground (unless you plan to connect the earth
grounded nodes together). The vast majority of Agilent power
supplies have outputs that are isolated from earth

This makes it sound like no big deal, but I'm a bit hesitant to put my DC supply in series with my function gen.

Edit: Using DG1032Z and DP832.

[HalfSpace]

Hi dadler,

Most function generators have a 50 ohm output impedance and having it in series with a power supply drawing a large current will damage the generator.

Try this link http://gpete.blogs.keysight.com/2012/09/simulating-power-supply-noise.html it may be the solution your looking for.

HalfSpace 

[Muxr]

Yeah I would definitely either build a simple noise gen as HalfSpace suggested, or get one of those single purpose generators from Ebay.

http://www.ebay.com/itm/DC-12V-Noise-Source-Simple-Spectrum-External-Generator-Tracking-Source-SMA-/251953200778?pt=LH_DefaultDomain_0&hash=item3aa994be8a

http://www.ebay.com/itm/ELECTRONICS-SALON-Pink-Noise-Generator-Module-Assembled-/201218811101?pt=LH_DefaultDomain_0&hash=item2ed99364dd

[edavid]

Most function generators have a 50 ohm output impedance and having it in series with a power supply drawing a large current will damage the generator.

This does not make sense.  The current is limited by the function generator output impedance.

What can possibly cause damage is if you apply an external voltage to the function generator output, e.g. if the load is shorted while the power supply is on.

[VK5RC]

The way I look at it the relatively high output impedance of the frequency generator will limit it's ability to put a low impedance 'noise'  on top of the DC signal, how about using a 'crappy" smpsu with lots of spikes (prob rel low impedance)  and filter them ' a bit'.

[HalfSpace]

Most function generators have a 50 ohm output impedance and having it in series with a power supply drawing a large current will damage the generator.

This does not make sense.  The current is limited by the function generator output impedance.

What can possibly cause damage is if you apply an external voltage to the function generator output, e.g. if the load is shorted while the power supply is on.

Hi edavid,

Depends on what the power supply voltage is set to and what the load impedance is. If we have a 30V supply and a 30R load and we insert in series a 50R function gen, apart from the load now only getting 11.25V across it the function gen has the rest. 18.75v squared divided by 50R equals 7.03 watts being dissipated in the function gen output circuit which could do it some damage.

HalfSpace

[alsetalokin4017]

Most function generators have a 50 ohm output impedance and having it in series with a power supply drawing a large current will damage the generator.

This does not make sense.  The current is limited by the function generator output impedance.

What can possibly cause damage is if you apply an external voltage to the function generator output, e.g. if the load is shorted while the power supply is on.

Hi edavid,

Depends on what the power supply voltage is set to and what the load impedance is. If we have a 30V supply and a 30R load and we insert in series a 50R function gen, apart from the load now only getting 11.25V across it the function gen has the rest. 18.75v squared divided by 50R equals 7.03 watts being dissipated in the function gen output circuit which could do it some damage.

HalfSpace

That's right. It pays to understand the output circuitry of your FG; some are a lot more robust than others. My old Racal-Dana / Interstate F43 has a couple of large-wattage resistors that establish the output impedance, so I can get away with series power connections that might damage more modern digital equipment. But in general, FG offsets can be established by connecting a battery or PSU in series with the FG's output and load.

The Agilent document attached below may be of interest.