Need 100V AC (1kHz-100kHz) low wattage source

Thread Starter

ponjavic

Joined May 4, 2011
37
I've tried to search some threads but this doesn't seem to be too common.

I'm trying to do electrophoresis (aqueous colloidal particles manipulated by a current).

For this I need to provide an AC source ranging from 40-100V with a frequency between 1-100kHz.

How can I get a cost effective source for this? I would assume that it is easiest to amplify the output of a function generator but where do I find a 100V source for the amplifier?

Googling gives me lots of results on audio amplifiers, would this be a possible route?

Currents are low. Ultrapure water 18.2MOhm/cm (5mm electrode gap). (Measured resistance is about 2MOhm)
 

t06afre

Joined May 11, 2009
5,936
What er we talking about here. Is it 100 Volt peak to peak, or RMS. or is it amplitude value?
 

t06afre

Joined May 11, 2009
5,936
Ouch square wave. That will require a bandwidht at least 10 times 100KHz. Or one MHz.
 

t06afre

Joined May 11, 2009
5,936
Also Electrophoresis, is the motion of dispersed particles relative to a fluid under the influence of a spatially uniform electric field. If you use a alternating voltage. You will get no netto movment. I think you want a pulsating DC swinging from zero to 100 volt instead of an alternating voltage
 

Adjuster

Joined Dec 26, 2010
2,147
A MOSFET gate driver IC working into a step-up pulse transformer might be one option, but such things can dislike inductive loads. The transformer would also need to have a nice low leakage inductance if you hope for good square waves at 100kHz.
 

Thread Starter

ponjavic

Joined May 4, 2011
37
A MOSFET gate driver IC working into a step-up pulse transformer might be one option, but such things can dislike inductive loads. The transformer would also need to have a nice low leakage inductance if you hope for good square waves at 100kHz.
Yeah the inductance was my reason for not going with a transformer.

As for the electrophoresis I do need AC. I should have been more specific, I am doing dielectrophoresis.
 

Kermit2

Joined Feb 5, 2010
4,158
maybe by following the links to the sources/websites listed in the article and blog you can find more resources to help you
 
Last edited by a moderator:

Thread Starter

ponjavic

Joined May 4, 2011
37
maybe by following the links to the sources/websites listed in the article and blog you can find more resources to help you
Good stuff I'll read into it.

Looks really dangerous :(
 
Last edited by a moderator:

Wendy

Joined Mar 24, 2008
21,839
Guys, we do not allow any power supply powered by mains here on AAC. The power supply that came up was 1000V, and directly connected to a 220VAC outlet.

In your eagerness to help please review what you are posting.

If the subject continues in this vein we (the moderators) are going to have to close this thread.

The OP mentioned a 100V P-P square wave. There are many ways to accomplish this.

A variation of this circuit for example.



It is possible to correct a 100V power supply with a isolated (this means transformer) power supply. Any other way puts people at risk. If you need something designed specifically for 100V let me know, and I will draw something up.
 
Last edited:

Kermit2

Joined Feb 5, 2010
4,158
While it could have been a mains powered circuit.

READING the post on the link would have informed you that the designer used an inverter to create the input AC. He knew about the danger of a mains connected circuit and this is brought out in the WORDS of the post there.
 

Wendy

Joined Mar 24, 2008
21,839
It does not matter, it was what it was, and the subject is not open for discussion. DC made directly from AC mains is a forbidden topic on this site, and is specifically mentioned in our Terms of Service (with my full and complete support) because it is a major safety hazard.

For future reference, an inverter for a direct AC to DC schematic is not valid, it is a way to try to get around AAC's rules. The schematic I saw when I opened the link did not show an inverter. It would have been in violation if it had showed an inverter, as it is a reasonable conclusion that you could eliminate the middlemen circuits and go directly to mains.

Building a 100VDC power supply is not difficult, but this was also a 1KVDC power supply instead of 100VDC. These types of circuits are not difficult to build. They look rather like inverters.

I suspect you will find the other moderators equally resolute on this subject. I am therefore requesting you to please drop this subject. The OP can achieve his goal without resorting to dangerous shortcuts. If the OP had brought it up this thread would already be closed with this notice...

Hello,

I am closing this thread as it violates AAC policy and/or safety issues.

Quote:
6. Restricted topics. The following topics are regularly raised however are considered “off-topic” at all times and will results in Your thread being closed without question:

  • Any kind of over-unity devices and systems
  • Automotive modifications
  • Devices designed to electrocute or shock another person
  • LEDs to mains
  • Phone jammers
  • Rail guns and high-energy projectile devices
  • Transformer-less power supplies
This comes from our Tos:
Terms of Service
Since he didn't I can only emphasize how serious we are about this. It is not safe.
 
Last edited:

Thread Starter

ponjavic

Joined May 4, 2011
37
Thanks for the input Bill. I guess it would be possible to build the circuit from scratch.

I am still wondering though if there is no simple way to get an audio amplifier to the the job with the output of my function generator.

I can easily lower my frequency requirement to say 1kHz, would that make things simpler?
 

Wendy

Joined Mar 24, 2008
21,839
The freq range is easy. A simple 555 could do it. They can even produce pretty decent symmetry.

If you were to use that circuit I showed with 90V zeners (or a combination of two zeners that added up to 90V) then you would have the oscillator you needed, all you would need is a power supply. The 4 transistor drivers is the thing, not the 555.

If you wanted to use a freq generator then have it drive a single common emitter transistor, then add something like the drivers I showed.

I will draw something up if you would like. What is your power source, wall outlet or DC?

A side note, the reason I was so harsh. The circuits shown in the removed links had a power plug. There was a verbal admonition to use an inverter in the plans, but give anyone a device that has a power plug and does what they want if it is plugged into the wall and it will be plugged into a wall socket. This is basic common sense. Moderators actually control very little, but what we can control we must. The disclaimer was to protect the author.

Lets start from the power supply. Back in the day of tube radios high voltage batteries were the norm. With some 9V batteries it is possible to interlock them to produce much higher voltages, 12 batteries will produce a rather unstable 108V.

If you want to do this with AC power making a simple transformer is not that hard, or even buying one for that matter. Making a DC power supply with batteries is easy too.

Varying the amplitude becomes the final issue. This is simple too. Basically you need to match a bunch of simple concepts into your final circuit. You mentioned low current output, how low current?

Here is one way to very output amplitude, there are others.

 

t06afre

Joined May 11, 2009
5,936
I think instead of looking for an analog amplifier. Why not take a look at the concept used for controlling DC motors the H-bridge http://en.wikipedia.org/wiki/H_bridge With that you can do with a single 100V power supply. And speed should me much more easy cope with.
 

Wendy

Joined Mar 24, 2008
21,839
Agreed. What I showed was a ½ H bridge.

With a full wave bridge you could use 50V, and have a 100V P-P signal.
 

Thread Starter

ponjavic

Joined May 4, 2011
37
Many thanks guys!

Yeah I would probably want to go for the amplifier route if possible.

My source is wall outlet but I wouldn't mind buying some cheap 100V power supply if they are available. Not really sure what to search for, most results come up as £300-400 proper sources which I think is a bit expensive. Since the current is so low I wouldn't mind going for the batteries either if I can cap them to make it stable enough.

The resistance is currently 2MOhm so not taking into account any dissipation (which I probably should with all the switching going on) for 100V, 50uA. In any case it should never reach a mA.
This might be completely wrong though as I believe water has capacitance so I'm not really sure how this affects the dissipation.

As my function generator is variable from about 1-10V would that not work for varying the final voltage?

A summary:
Need to find a fairly cheap 100V supply or build it with batteries if that is feasible
Amplify the signal from the function generator with a ?common emitter amplifier?
Do the amplification using a full wave bridge

I guess my biggest issue so far has been getting the 100V supply. The amplification and the H-bridge design really should not be an issue if the current is as low as I expect and if the capacitance of the water is not a major issue.
 

Wendy

Joined Mar 24, 2008
21,839
I would state, for a square wave, a 1V Peak wave would work.

Have you figured out the power supply yet (let me guess, you haf a plan!). :rolleyes:

Are you wanting to go with the H bridge technique? Remember, the power supply can be half the value (50V).

If you make a variable voltage on the power supply that will also vary the P-P voltage on the output.
 

Thread Starter

ponjavic

Joined May 4, 2011
37
I would state, for a square wave, a 1V Peak wave would work.

Have you figured out the power supply yet (let me guess, you haf a plan!). :rolleyes:

Are you wanting to go with the H bridge technique? Remember, the power supply can be half the value (50V).

If you make a variable voltage on the power supply that will also vary the P-P voltage on the output.
I haven't figured out the power supply :) also I want 200V p-p so I get +-100V with the H-bridge.

Basically would the batteries be feasible for this application? If so I'll just go with that and stack them. Could I then use the circuit you posted (but replace the 555 with my function generator)?

Any particular transistor I should be looking at? I.e do I have to make the bandwidth 10x my actual frequency requirement or something like that?
 

Wendy

Joined Mar 24, 2008
21,839
Not really. I'm about to leave, so I will get back with you later.

The H bridge has an odd feature. It will allow you to make apparently twice the P-P voltage as the power supply. Look at the link that was provided and see if you can figure out why.
 
Top