How to superimpose 15V AC onto a 5000V DC

[[email protected]]

Hi all,
Please tell me how can I safely superimpose a 15Volt Ac signal (50Hz -
100KHz) from a signal generator onto a variable DC voltage (50V to
5000Volt). The output is 5000V DC + 15 V AC.

First I thought i wil use a transformer to isolate the function
generator output from the HV and then use a diode to block Ac from
getting into the HV power supply. But this can remove only a half cycle
of the Ac pulse and may damage the HV output.

Please let me know if anyone has a nice idea or some circuits.. I am
very much in need of this circuit...thank you.

regards,
kishore

 

[Luhan]

Hi all,
Please tell me how can I safely superimpose a 15Volt Ac signal (50Hz -
100KHz) from a signal generator onto a variable DC voltage (50V to
5000Volt). The output is 5000V DC + 15 V AC.

An you want to do this for what reason?

Luhan

 

[[email protected]]

hi luhan,
thank you for ur reply.
I need this for studying material properties, orientation of molecules
under a static+oscillating electric field.
do u have any ideas ?

Kishore

 

[Luhan]

hi luhan,
thank you for ur reply.
I need this for studying material properties, orientation of molecules
under a static+oscillating electric field.
do u have any ideas ?

I did something like that once with 700 volts of audio riding on an 800
volt dc source. This was used to create audio from 2 electrodes placed
in a candle flame.

It all amounts to the insulation resistance of the transformer feeding
the audio. The primary/secondary insulation resistance must be able to
withstand the high DC voltage. If so, you just connect the tranformer
secondary in series witht the high voltage source.

Luhan

 

[[email protected]]

But, will this damage the output of the HV power supply ? as the
secondary generates AC, it will go in to the DC unit, even if it is in
series.

I planned to have a transformer with primary turns and sec turns of the
same number so that no voltage amplification takes place. But stil the
above problem is there..

 

[[email protected]]

But, will this damage the output of the HV power supply ? as the
secondary generates AC, it will go in to the DC unit, even if it is in
series.

Conceptually, you have to ask the question of how the DC power supply
would even know that you added this AC voltage to it's output?

The only way it can know is if either your transformer isolation breaks
down (hope not) or if there is a complete circuit which allows current
to flow through the supply.

Consider what the added AC component can do to change the current flow
through the power supply. Can it make the current flow larger? Yes.
Can it make it smaller? yes. Are the degrees to which it can do so
within the current rating of the power supply?

The important question: can the AC component make the current flow
negative - can it make current go through the power supply the wrong
way? It would seem it can only do this if the peak AC voltage is
higher than the output voltage of the supply. Can that situation
occur? Sure, when the supply is turned off. Is it a problem? All
depends on how the supply is built.

But if you put in a diode at the power supply output, then current can
only flow through the supply in the designed direction. However, if
the AC voltage ever approaches cancelling out that of the supply such
that the diode comes close to cutting off, then your AC waveform will
of course be distorted - but distorted signals and equipment damage are
two different classes of fault.

 

[Tim Wescott]

But, will this damage the output of the HV power supply ? as the
secondary generates AC, it will go in to the DC unit, even if it is in
series.

I planned to have a transformer with primary turns and sec turns of
the same number so that no voltage amplification takes place. But
still the above problem is there..

I doubt that a 5kV DC supply would notice the 15VAC. You'll be lucky
not to have 15V of ripple on the DC supply's output, at no extra charge
to you.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html

 

[John Fields]

But, will this damage the output of the HV power supply ? as the
secondary generates AC, it will go in to the DC unit, even if it is in
series.

I planned to have a transformer with primary turns and sec turns of the
same number so that no voltage amplification takes place. But stil the
above problem is there..

---
Please bottom post.

If interaction with your supply is going to be a problem, isolate
your supply with a resistor:


HVDC
|
[R]
|
ACIN>----+ +
P||S
R||E
I||C
ACIN>----+ +
|
|
[YOUR EXPERIMENT]
|
DC GND


Also, if your supply is fast, you might be able to able to control
its regulator with your AC signal and avoid the whole transformer
issue.

 

[Genome]

hi luhan,
thank you for ur reply.
I need this for studying material properties, orientation of molecules
under a static+oscillating electric field.
do u have any ideas ?

Kishore

You might assume, hopefully, that the sample you are examining has a very
high DC resistance. I assume it is placed between a pair of plates to which
you apply your supply. Once these have been charged then effectively no DC
current will flow.

When you apply your AC voltage, through the transformer, then much of the
current flow will be capacitative and determined by the frequency and the,
separation and area of the plates along with the dielectric properties of
your material.

Your power supply will have a smoothing capacitance across its output
terminals and, as long as the value of this capacitance exceeds that of your
plates by a suitably large margin then the AC current will flow through that
capacitor and bypass any active parts actually within the supply.

Things should be safe without have to add any further components.

DNA

 

[Joerg]

But, will this damage the output of the HV power supply ? as the
secondary generates AC, it will go in to the DC unit, even if it is in
series.

Not likely. What AC frequency?

The main issue will be a sufficient breakdown voltage between
transformer primary, secondary and core. Some medical grade iso
transformers go that high. If you need 50/60Hz here you might want to
talk to Ulveco or companies like that.

Also, if using 50/60Hz keep in mind that what comes out of a wall outlet
is hardly ever truly sinusoidal.

 

[Phil Hobbs]

Not likely. What AC frequency?

The main issue will be a sufficient breakdown voltage between
transformer primary, secondary and core. Some medical grade iso
transformers go that high. If you need 50/60Hz here you might want to
talk to Ulveco or companies like that.

Also, if using 50/60Hz keep in mind that what comes out of a wall outlet
is hardly ever truly sinusoidal.

Some HV supplies have floating outputs, or they can be made floating
trivially, e.g. by removing a strap. Putting a transformer between the
cold end of the supply and ground would relax a whole lot of these
difficulties, and if correctly done, would be somewhat safer. Do ground
the core of the transformer, and put some transient absorbing Zeners
across its output just in case something opens up without your knowing it.

Note that this suggestion assumes that you know enough not to kill
yourself doing it. If the output of your power supply can supply enough
current to kill you, don't mess around inside it unless you *really*
know what you're doing. Some supplies do, some don't.

Cheers,

Phil Hobbs

 

[Joerg]

Hello Phil,

Note that this suggestion assumes that you know enough not to kill
yourself doing it. If the output of your power supply can supply enough
current to kill you, don't mess around inside it unless you *really*
know what you're doing. Some supplies do, some don't.

Absolutely, especially if there are capacitors involved. For example,
even a wimpy power supply can lead to cardiac arrest and possibly death
if it was used to charge up one of those microwave capacitors.

 

[Ross Herbert]

Hi all,
Please tell me how can I safely superimpose a 15Volt Ac signal (50Hz -
100KHz) from a signal generator onto a variable DC voltage (50V to
5000Volt). The output is 5000V DC + 15 V AC.

First I thought i wil use a transformer to isolate the function
generator output from the HV and then use a diode to block Ac from
getting into the HV power supply. But this can remove only a half cycle
of the Ac pulse and may damage the HV output.

Please let me know if anyone has a nice idea or some circuits.. I am
very much in need of this circuit...thank you.

regards,
kishore

You might like to do some research on analog telephony principles
since this function is commonplace in that industry. As an example,
the ac ringing potential (nominally 75Vac at 20Hz) is superimposed on
the -48V dc (wrt +ve gnd) line potential during ringing.

 

[legg]

Hi all,
Please tell me how can I safely superimpose a 15Volt Ac signal (50Hz -
100KHz) from a signal generator onto a variable DC voltage (50V to
5000Volt). The output is 5000V DC + 15 V AC.

First I thought i wil use a transformer to isolate the function
generator output from the HV and then use a diode to block Ac from
getting into the HV power supply. But this can remove only a half cycle
of the Ac pulse and may damage the HV output.

Please let me know if anyone has a nice idea or some circuits.. I am
very much in need of this circuit...thank you.

It's easier to 'superimpose' the HVDC signal on top of the LVAC
signal, using a grounded transformer, or a simple voltage source in
the HVDC return path.

RL

 

[Frithiof Andreas Jensen]

Hi all,
Please tell me how can I safely superimpose a 15Volt Ac signal (50Hz -
100KHz) from a signal generator onto a variable DC voltage (50V to
5000Volt). The output is 5000V DC + 15 V AC.

Float the supply on the 15 V AC - i.e. the 15 source is in series with
the return of the 5000 V supply. Much easier than doing it the other
way round

 

[Kishore]

Float the supply on the 15 V AC - i.e. the 15 source is in series with
the return of the 5000 V supply. Much easier than doing it the other
way round

Hi all,
Thank you very much for the solutions and ideas. I think i wil use
Frithiof's suggestion. I hope my Stanford Power supply won't mind
injecting an AC to its neat clean DC output.
best regards,
Kishore

 

[Mark]

Hi all,
Thank you very much for the solutions and ideas. I think i wil use
Frithiof's suggestion. I hope my Stanford Power supply won't mind
injecting an AC to its neat clean DC output.
best regards,
Kishore

Agreed, injecting the AC in series with the gound side of the HV supply
is a good idea... BUT BEWARE OF THE FAULT CONDITIONS...if something
faults open on the ground side, all the wiring on the high side of the
fault assummes a high voltage. This is dangerous because a wire that
is normally near ground and not well insulated and not well respected
can unknowingly be at a deadly high voltage. So put a resistor or
zener or other device to limit the voltage that can appear on the
"ground " side in case there is an open circuit fault.

Mark