Variable frequency sine-wave AC (up to 80,000Hz) and variable voltage (up to 2,500V) bench generator needed?

Question

I am making a very special type of lamp that is low current (up to about 10mA) that is AC powered, between 40,000Hz and 80,000Hz. The lamp voltage, depending on the length, is between 1,000 volts and 2,500 volts.

I am looking for some type of bench generator that I can input for variable parameters of freq, voltage and current for testing the characteristics of different lamps.

Any thoughts?

Thanks, Bob
totallylamps@ yahoo.com

Answer 1

I doubt you will find anything on the market, you will have to build your own.



The frequencies can be gotten from a garden variety function generator or sine wave generator.



It's generating the 2500 volts at 10mA that is difficult, as power devices don't go much above 1000 volts. The power involved is 25 watts, not too high.



If you can find or make a step up transformer that will work at those frequencies and voltages, then a good audio amplifier will handle those frequencies and powers ok, the output voltage is just too low, and a step up transformer will fix that... just an idea...

Answer 2

Agreed, this is a tough one and I actually also question the need. Very curious about the lamp and as to what might be its normal power source??? How do you light it up now?...You may need to modify that power source, if it exists, to have a variable frequency.



Again, the frequency source is not the problem, it is the amplifier to produce the voltage and current. You are well above audio frequencies, but well below RF. You seem to be in the ultrasonic range.



Suggestion might be to look for some Sonar related test equipment, perhaps Military surplus. Trouble is I doubt the voltage requirement will be there.



The other place to look may be some of the vibration test labs. Although these frequencies may be a bit higher than used in that field, they may have something. There are amps and frequency generators in that field that essentially drive vibration tables like driving a high powered speaker to generate controlled vibrations.



Not sure if there is anything at this web site, but check out http://www.pcb.com/ these guys work in the vibration and instrumentation field and may be have some info, at least as to a direction to go in.



The other possibility is the test equipment used to do voltage breakdown testing of insulators. I recall using a device called a hypot. Don't know if the new stuff gets up that high in frequency, but it very well might. It puts a variable voltage at some variable frequencies on a device to find out at what point the insulation breaks down and forms a hazard.





Other than that, I do believe that this will end up being a very customized piece of test equipment.





Good luck!..unfortunately, I think you will need it!

Answer 3

It's actually very simple. If there was no host, and only two doors left, of course, the new odds would be 50/50 (like in a coin toss, the previous flip has zero effect on the next flip- even if it came up heads 20 times, the odds are still 50/50). I don't care about 33% and 66% and how many doors there were. If at any point in time, there are only 2 doors left and you pick 1, your odds all day long are 50/50 IF THERE ARE NO OTHER VARIABLES. In this specific case, there is one additional and highly significant variable at play- THE HOST. Predicated upon the assumption that the host knows exactly where the car is, there is a statistically significant probability that it would be in the hosts best interest to have the player switch. Knowing that the player has the winning hand, the host is therefor motivated to have him change doors (change his "original decision.") If his "original decision" was incorrect, by contrast, it would be in the hosts interests to allow him to keep his original choice of doors. In this case, he simply offers him a choice that statistical research has proven, the player is likely to accept and will increase the odds of the host. Gambling is very much a science and in the long run, the house almost always wins. It's like in a coin toss, if the other person chooses heads and you peak and see it's heads, you might say, "are you sure?" It's really not rocket science but an interesting concept to think about on a rainy day. It's also a nice way to set up the movie as it provides contrast and irony to the plot but I digress- getting a bit off topic!

Answer 4

Yes. Bench generators' outputs are in the low volts, low watts range. 2500 volts at 10 mA is 25 W.



Your requirement is so far out of the ordinary that you are not going to get it in one instrument. The function generator part is run-of-the mill. The power amp, on the other hand, would be a very special custom design. I would want to know more about the application to believe it's justified.

Answer 5

Thanks, Gentlemen. I am actually trying to design a very special type of EEFL (external electrode fluorescent lamp) that runs at about 6-8mA... between 60,000 - 80,000Hz and depending on the length of the lamp, between 1,000v and 2,500 (sine wave). So every lamp I make has different characteristics.



I need to find the optimum operating parameters, hence the need for the "variableness".



Thanks,



Bob