The little backlight lamps I have used run at a couple of hundred volts. The inverters are basically AC constant current sources. An oscillator at ~20kHz drives a high voltage transformer producing a couple kV AC at a few mA. This is fed out via a low value capacitor, so there is kV available at the output to strike the lamp when it is high impedance, then operates as a constant current drive when it is low impedance. But the transformer still maintains most of its original voltage, it is dropped across the series capacitor.
This is for the small CCFL invertors used for LCD displays.
But Phils diode sees the full transformer reverse voltage, potentially several kV. A bridge rectifier would avoid this.
They can easily supply a few mA but they aren't really well regulated, at least not at low voltages. Such converters can be modified with a current source in the central primary connection but at very low output voltage levels they can behave more like a two-stroke engine with fouled spark plugs. Not very clean, if that's important to Marco.
I do a lot of low power, on-chip, stuff. For your sort of requirement think "energy-per-unit time". Then burst-mode flyback with added variable "charging" current control will make a nice mathematical exercise ;-) ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
It's basically a constant current source, which is why I suggested a shunt regulator. A bridge rectifier would be a quite reasonable addition.
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics
160 North State Road #203
Briarcliff Manor NY 10510 USA
+1 845 480 2058
hobbs at electrooptical dot net
http://electrooptical.net
I'd use a step-up switcher. Maybe one with a gapped transformer to keep the duty-cycle reasonable and then use a linear regulator to make the required voltage. Some transistors and an opamp should do the trick.
--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
The terms are fuzzy, so you need to use your judgment, and someone will tell you you're wrong no matter what.
Where you put your eyes, mostly, but a power supply is usually something that has to deliver power to a load whereas a signal generator is something that makes a signal (not just a DC level). If the characteristics of the signal are important and the load is negligible, then it's more on the signal generator side. If the "signal" just needs to be constant DC and how it gets there isn't important then it's more on the power supply side (or bias generator?). If it is actually powering something, however small that power may be compared to the power consumption of the rest of the circuit, then it's a power supply.
Somewhere in the middle would be a "bias generator", if it just needs to put 300V at some circuit point but doesn't need to actually deliver power. Once again, the term is fuzzy.
Having said all that, if there is a part of your circuit that actually does something, and does it with 300V, then no matter how small the current is you're building a power supply.
--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com
The 0 to 300V will be delivered to an actuator (I don't know what it is at the moment). It uses this "signal" (in my handbook! :) ) to do something. In fact, as I've already said the voltage will be controlled by an MCU, say the output of a PID.
Then perhaps "amplifier" will better capture the usual qualities that you want it to be able to live up to.
If this is going to be inside a control loop then bandwidth and linearity (or at least easily quantifiable and predictable nonlinearities) are both significant features -- and a big part of why I'm now trying to rename it as an "amplifier".
Also, since you're looking at amplifier-like operation, one needs to ask how many quadrants of operation you need. You've specified 0-300V, so that relieves your circuit of responsibility for two out of the four -- but is it always ever going to just source current to the actuator, or is it going to need to suck charge out of the actuator at times? If your amplifier needs to sink as well as source current then its a two-quadrant device. (Certainly if your actuator is a piezo-electric one you're going to be kissing large-signal bandwidth goodbye if you don't use a two- quadrant amplifier).
If you need high-bandwidth, two-quadrant operation, then a higher-than
300V DC supply plus a linear amplifier is going to start looking really attractive, at least as long as it doesn't add too much current consumption for your overall power budget.
But since you're not sharing your bandwidth and linearity needs its impossible to judge whether you can meet them with just a traditional switching supply topology and control circuitry that holds its mouth right, or if you need to augment that with a linear amplifier circuit.
I think you really need to share your bandwidth and linearity requirements with us; if you don't know enough to say then describe what you're doing and we'll give you various wild-ass guesses as to your needs.
A good rule of thumb for component bandwidths is to multiply your loop bandwidth by ten and try to exceed that everywhere (this is where the "sample at 10x your bandwidth" comes from). Amplifiers are generally the easiest things to push above that figure, actuators are often whatever you are stuck with.
To determine your loop bandwidth, determine your desired settling time for your loop and take its reciprocal -- that's going to be close enough as long as the loop remains in linear operation throughout its whole trajectory.
If your loop is not going to be operating substantially linearly -- most typically if some element in your loop is going to be rate limited -- then you need to take the total settling time _after_ it's gotten close enough to the target to be in a linear range and take the reciprocal of that -- it will, of course, be significantly higher.
--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?
Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
Oops -- I forgot to mention. The title of this thread is far enough away from a "0-300V amplifier with a 24V supply" that you may just want to start a fresh new thread, making sure to point back to this one.
--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?
Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
Would it be evil to ask if you can recommend any chips?
--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?
Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
A C-W has a lot of losses from diode drops and cumulative capacitor impedances.
24 volts p-p to 300 DC would take a heap of stages. A little flyback converter can be made with a cheap Digikey inductor or tapped inductor, followed maybe by a dual-diode doubler thing.
High boost ratios are hard for a C-W and for a single-inductor boost converter. In the single-inductor boost case, at high boost ratios, a lot of energy goes into charging the fet capacitance and the coil parasitic capacitance. 10:1 or maybe 20:1 boost get tricky. That's where a tapped inductor or transformer get useful.
Lots of people make dual-winding inductors, configurable as a center-tapped inductor, which is handy in a low-power boost application like this one.
formatting link
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
One of my custom chip designs ?>:-} I've done a number of designs for cell-phone displays or LED drivers that worked that way.
I don't do discrete stuff, I only do device-level designs, so I'm woefully unfamiliar with off the shelf chips. But I'd guess your search terms should be "Linear flyback current-mode", Linear Technology seems to specialize in that sort of thing.
Actually got a call on Thursday from a potential client looking for an analog designer. His complaint was that all he could find were analog "designers" who could only thumb thru catalogues (plug-and-play) >:-} ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
We made and sold exactly that. It was only two stages and put out clean voltage at better than the rating he mentioned.
Very clean, in fact. Flybacks are very noisy for this application.
I have a schematic here somewhere. They no longer sell the product.
2 stages and final output storage. Gets pumped by an oscillator and a small, precision tuned gap transformer front end.
And C-W losses you speak of are related to HV designs where the diodes are in fact multi-element (in the same package) diode stacks. They are also negligible (the voltage drops) after the first stage of any well designed c-w array.
This design can use a much lower voltage diode. Single element even.
Oh and your "dual diode doubler thing" IS a single stage c-w. Duh. Much better to pump this thing with something closer to a sine wave. Far lower noise specs in the output.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.