200W Adjustable Voltage Regulator and Power Amplifier circuit / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

200W Adjustable Voltage Regulator and Power Amplifier circuit

Move Thread

LAN_403

Uriah

Thu Apr 03 2008, 12:05AM

Registered Member #898 Joined: Sun Jul 15 2007, 07:50AM
Location:
Posts: 20

I am looking for a solution for a 200Watt 12V amplifier adjustment stage. This amp will switch at a specific sine wave frequency (47kHz +/- 1.5kHz) with adjustable duty cycle and peak voltage. Right now I can adjust the voltage using LM317 in 2mV steps, but the max current for LM317 is 1.5A. These units are not very efficient either. I am using a single DirectFET to switch 12V up to 290A, so the voltage adjustment must be between the PWR source and the DirectFET Positive Emitter.

What is the best way to adjust voltage in this way at 12V with a 30A maximum? Should I use multiple Positive Adjustable Voltage Regulators in parallel? Optimally I am looking for a solution that is over 95% efficient, even if this requires serious cooling.

My design requires a DC Voltage Regulator that can maintain 12V with an adjustment feedback circuit to raise the voltage in 2mV steps. The amplifier is driving twenty multi layer piezoelectric transformers (10W each) whose max output is 3.46kV. This output is rectified to DC and through the use of a voltage divider and current shunt sensed and corrected. The voltage reads out in mA instead of kV (3.46kV = 346mA). Voltage output from the Piezo Transformer drops over time, that is what the feedback and sensing is to control by slightly increasing the input V.

Best regards,
Uriah

Electroholic

Thu Apr 03 2008, 12:55AM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

if it is 95% efficient, it won't require much cooling.
Efficiency of linear regulators can be calculated as Vout / Vin. And Vin min is usually a few volts above Vout, for the 78series anyways.
So you probably want a buck regulator.

346ma At 3.46kV? that is like 1.2kW, i thought you only got 200W worth of transformers.

HV Enthusiast

Thu Apr 03 2008, 02:44AM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Uriah wrote ...

I am looking for a solution for a 200Watt 12V amplifier adjustment stage. This amp will switch at a specific sine wave frequency (47kHz +/- 1.5kHz) with adjustable duty cycle and peak voltage. Right now I can adjust the voltage using LM317 in 2mV steps, but the max current for LM317 is 1.5A. These units are not very efficient either. I am using a single DirectFET to switch 12V up to 290A, so the voltage adjustment must be between the PWR source and the DirectFET Positive Emitter.

What is the best way to adjust voltage in this way at 12V with a 30A maximum? Should I use multiple Positive Adjustable Voltage Regulators in parallel? Optimally I am looking for a solution that is over 95% efficient, even if this requires serious cooling.

My design requires a DC Voltage Regulator that can maintain 12V with an adjustment feedback circuit to raise the voltage in 2mV steps. The amplifier is driving twenty multi layer piezoelectric transformers (10W each) whose max output is 3.46kV. This output is rectified to DC and through the use of a voltage divider and current shunt sensed and corrected. The voltage reads out in mA instead of kV (3.46kV = 346mA). Voltage output from the Piezo Transformer drops over time, that is what the feedback and sensing is to control by slightly increasing the input V.

Best regards,
Uriah

Well, first of all, you ain't going to get even remotely close to 95% efficiency using a linear type regulator especially at those power levels.
Whats the input voltage? And what is the output voltage range?

...

Thu Apr 03 2008, 02:56AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

I think he is saying that he has some kind of amp in his sense circuitry, so he reads 300ma for 3kv output, despite the fact that there is only 60ma of current being produced by the transformer (although you never know...)

In any case, what is the input voltage? You best bet is probably a boost converter, although it will require careful selection of the filter caps and inductor to be able to produce a 45khz sine wave on the output...

Short of that, it is possible to wire external pass transistors to a lm317 so that it would be able to handle the large amount of current--just keep in mind that you are going to be burning a few hundred watts in their heatsink. Also, if you are going to be adding external transistors you may as well use a lm723 since all of the power will be dissipated in the external transistors.

Uriah

Thu Apr 03 2008, 05:59AM

Registered Member #898 Joined: Sun Jul 15 2007, 07:50AM
Location:
Posts: 20

Sorry for that typo, I meant 3.46kV reads out as 346mV, not mA. The current output from the collective Piezo Transformers is 57.8mA (with losses around 56.64mA = 196W max). I was referring to the voltage divider (10,000:1 ratio) so that I can more accurately measure the HV using a DMM in the hundreds of mV range. Even if I read the voltage in volts (3.46V) the inaccuracy would be between 1-2%. in the mV range accuracy becomes irrelevant. My DMM can chart 1000 readings per second. Although using a high resistance divider such as this will produce noise in the output. Readings will be made every minute for one pulse width period, to calibrate the input via the Voltage regulator.
Thin input V into the REGULATOR is 12V which is the normal minimal output anyway. Some loss is still attributed to heat loss here, correct? I am thinking the 12V line can be switched to bypass the voltage regulator until it is sensed to step up the voltage. Basically I need an adjustable step-up voltage regulator with an input of 12V and an output between 12V-13V (2mV steps).

What is the nominal efficiency of a buck converter? In fact DirectFETs are already used in buck converters for processors.

Best regards,
Uriah George

HV Enthusiast

Thu Apr 03 2008, 02:45PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

wrote ...

Thin input V into the REGULATOR is 12V which is the normal minimal output anyway. Some loss is still attributed to heat loss here, correct? I am thinking the 12V line can be switched to bypass the voltage regulator until it is sensed to step up the voltage. Basically I need an adjustable step-up voltage regulator with an input of 12V and an output between 12V-13V (2mV steps).

Uh, no its not. No linear regulator in the world is going to give you a 12V output with a 12V input. The LM317 you mentioned requires at least 2V headroom i believe, so you need at least a 14V input.

I'd recommend a little more researching on linear regulators in general before continuing with this thread. Check out wikipedia on linear regulators, or just read some of the LM317, 780x datahseets, to become more familiar with their operation.

For example, if you are using an LM317 to regulate 12V output, you need at least 14V input voltage. The power dissipation in this device is 14V-12V * current. If you are drawing 1A, then you have 2W dissipation. This would give you an efficiency of 85% maximum (approx)

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob