sine wave current booster / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

sine wave current booster

previous 1 2 3 4 next last

Move Thread

LAN_403

Hazmatt_(The Underdog)

Fri Aug 01 2008, 07:39AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Yea I had to change those out to 1500R because they were dissipating too much heat. Basically what they are doing is setting the bias current into the transistors. Your current gain is = B (beta) and if you want a particular collector gain of say 100, then Beta needs to be 100, and Ib X Beta = Ic, so you need 1/100 current flowing into the base in order to achieve that current gain.

WOW! Nice Output! ~8W is a lot more then what I'm playing with right now! I'm Impressed :)

I'm going to try to get something going for the range you want because my needs are similar. So I'm still looking at transistor specs and I'm going to try to get to ~100W but that will likely require MOS, so I'm going to try to go that direction as well. I need similar outputs so we can sorta work on this together, heheh. But that's great news!

Now don't be surprised if you need relativily high voltage with BJT's to achieve your goal. They may require +- 50V supply at a couple of amps to deliver the response necessary, and MOS of course will need maybe 20V but at very high current.

Hope to get more going soon, I have to wait until Sat. for my time off period to start. but some things should result soon. :)

Steve Conner

Fri Aug 01 2008, 11:33AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hazmatt_(The Underdog) wrote ...

WOW! Nice Output! ~8W is a lot more then what I'm playing with right now! I'm Impressed :)

Yeah, that's what he gets for using 100 ohm resistors and putting up with the extra heat.

Hazmatt_(The Underdog)

Sun Aug 03 2008, 11:55PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

I'm going to try to get this working with a mos push pull driver, but the consequence with that is that no its not going to be direct coupled output so I can't push the response down below ~10KHz without getting a special core. I still havent received my op-amps yet so I can't go from DC to 160MHz, but I do have a phase splitter prototyped from ~8KHz to 19MHz :) Below its lower cutoff it won't do anything, and above that the phase misalignment is bad enough that the devices should have power problems.
It's fairly flat response in its working range, which is good, along with a good stable phase response. My splitter here has isolated outputs incase I have to use different push pull configurations.

The resistor across the transformer is necessary to normalize the transformer response. Other values may optimize the transformer for better performance, but this is what I had on hand. I think the driver does quite well for a TIP102 considering this is only drawing 50mA at regulated 12V.

Here's some info:

Pictures showing phase splitting at ~8Khz, phase alignment to 19MHz and phase misalignment at 24MHz

hard to believe this goes to 19MHz huh!

Turkey9

Mon Aug 04 2008, 11:41PM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

Wow thats great. Just one question, what are the turns on the transformer? is each section 12 turns? i can't quite understand that part of the schematic.

Hazmatt_(The Underdog)

Mon Aug 04 2008, 11:54PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

It has 3 windings ( Trifilar ) each is 12 turns.

I'm almost there, but I'm having a terrible time getting the power amp to function properly. It doesn't want to do squat in the linear mode. I was playing with it in switched mode and got really ugly waveforms, but the impulse was enough to base drive a secondary coil to the point of 2cm discharge, so its getting there.

I'm going to have to experiment a lot with turns ratios, core materials, and mosfets before this problem gets close to an answer. For your freq range and mine we need a kind of powdered metal core. The color you'd be looking for is grey. I have a couple of cores that are close.. grey/yellow, so I'm going to try those insted of the ferrite. I don't have much hope... but I'm trying.

Matt

********************************** ******************

Been trying to get a BJT linear amplifier going but there's a problem, the simulation works but the current mirrors in real life do not. I've been having terrible luck getting the output stage to produce any real power. I've burned up a couple of 100R 2W loads, but the problem is the voltage amplifer stage, I just can't seem to get it to amplify enough for the power stage.

I've built a current source so I can track current settings, setup several voltage amplifier stages with little luck and I'm not totally sure the schematics I have even work. So what I'm going to do is build one straight out of the book and see if the stupid thing works. If it does, then I'll go back and revise the OPA so it will handle your freqency response.

That's about all I can think of, because this is really a first for me here.

******************************************* *************************

I've been playing with this and it seems to work well. A couple of notes though,
1. The diodes should be replaced with schottky's for better performance
2. I tested it with a dome lamp for a car and it will work to loads of 1 ohm or a bit less (as the lamp glows)
3. I did test it with +78 and -78V supply and it works, but the transistors start getting pretty warm after a short time, I don't recommend going over +50V and -50V. The higher the voltage the better because that will keep the sine wave intact. I did notice a little step formation at the +38V and -38V supply due to the diode, hopefully the schottky will help with that.
4. The amp requires a pretty large input voltage, my supply only goes to 7.5V for a 10V output of the amp, so your gain ratio is 10/7, not great considering the amount of input power required to get this thing going, but then again its a current amplifier not really a voltage amplifier.
5. If you want your 20A spec. I could see doubling up the transistors for that job, shouldn't be too hard to push some real current.
6. What I really forsee is putting an opamp in front of the power amp to gain up the input signal to 12V RMS and hopefully increase the gain. Ive only got up to 7.5V RMS, so 12 or 14 should really be interesting. After that the output would be taken and fed back to the input opamp with a conditioner to treat the output as a reference level, then form an automatic gain control which would be really sweet for a hf power amp like this is beginning to be.

So here it is:

*********************************************** ***********

Wellp... I've tried my darndest to go direct coupled without any boradband transformer or the like but I'm simply fed up with the preamplifier. If I knew a good opamp to mate to the power amp I would do that in a heartbeat, but the trouble is I don't know the HF stuff all that well, never covered it in EE, I have a bunch of books on it and they're no real help, my audio amp stuff will not work at all, and I'm getting fed up.

I ordered some 15MHz opamps but I have learned that they are not worth a shit. They will not do 4MHz, they will not go rail to rail, and they are totally useless. So I'm at a loss right now and I need a buffer amplifier or something, but I don't know what.

Anyway, I hope to finish this soon so I can get on with testing.

**************************************** ***********************************

9-7

through careful observation I actually got some response from a class B driver of sorts, it resembles the voltage amplifier stage seen in most audio amps with the exception that it still uses the two diodes rather then the amplified diode and the current source. I may look into the current source and diode in the future, but for now what I have works. I will post what I can about it later, but the important bit is that I have a front end that is DC coupled and will operate from ~50Hz to ~5MHz. The power output does vary, but the amplifier does output a decent amount of power, not 100W, but decent enough to drive another power stage or a cascade of power devices.
I'm pretty happy with what I have here at the moment, but I would like to extend the range a bit more so I'm going to test this out with some power transistors that are specd. to 20MHz. Hopefully along with the different transistors I can implement automatic gain control to flatten out the amplitude and turn the amplifier into a decent instrument.
I also want to post some more pictures once I get the camera back from my dad.

Hazmatt_(The Underdog)

Sun Oct 05 2008, 04:19AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Almost there!! ^^

I was attempting to get the 10MHz mark with the "power brick" on the left but it won't validate past 100KHz even though the transistors spec. to 20MHz. They should go to at least 2MHz before they crap out.

The funny thing is the class-B amp on the right uses transistors that are rated for at least 3MHz and I can push them to 5MHz with little amplitude degredation, and to 7MHz with some noticable degredation.

Your current booster will work fine with a +- 40VDC supply powering NTE 87's and NTE 88's. So its almost there. ( I have made my decision to go with these ones since they work pretty well.)

I think transistors are funny things in this regard because some may work, some may not, its hard to tell until you sit down and validate them for your application. As far as your current needs are concerned, I do not know as of yet how much current this amplifier will supply into a load, that still has to be tested, but what I want to do with this amp right now is split the 4 power transistors into 2 heat sinked units to dissipate the heat better and then build in an AGC unit to stabilize the amplitude. Once that is done I will move on to temp. monitoring, input amplitude monitoring, output amplitude monitoring.

I'm pretty satisfied with what I have here for now, so I'm going to develop this amp further, and then work on more powerful amplifiers soon.

Steve Conner

Sun Oct 05 2008, 01:00PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hint, Zobel network on the output and stopper resistors on your NTE87/88 bases will stop any parasitic oscillations you may get as you turn up the beans. If you really want a lot of RF power, metal-gate MOSFETs are hard to beat.

Also, you probably ordered the wrong op-amps. A 15MHz GBW does not mean it does anything worth a crap at 15MHz. My colleague wanted to build an amp with a gain of 10 at 44MHz, and he used an op-amp with a 1.8GHz GBW. It worked fine. I've used the AD811 to drive 50 ohm loads up to 40MHz. Two of them in parallel for extra power.

Hazmatt_(The Underdog)

Sun Oct 05 2008, 07:51PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Yea I neglected 20dB/decade rule when ordering that amp. The 15MHz one should only give decent gain to 1.5MHz.

I got some 100MHz amps ordered and they're here now so what I'm going to do is basically copy an existing design to get the ~1W output drive power with the parts specd. to 100MHz.
The hard part for me is getting that 1W boosted to 5W to drive the real power transistors, but I am making an effort of that. The unfortunate problem is that the books cost as much as the transistors O.o, and I have neither really since I'm a controls guy not RF & Comm. So I'm slowly amassing the books and parts to do this.

The op-amp book helped because it pointed me in the right direction again, and I should order some 1.8GHz op amps too.

I was working with mosfets for a while but the trouble with mosfets is they like to latch up, square your sine wave, and generate lots of harmonics. The real problem this causes me is that I need to make measurements from the output of the amp so wave purity is a concern. My situation is either get the transistors working in a decent amp, or tubes. My tube amp is almost workable but its phase splitter is badly out of alignment so I have to scrap the front end.

Anyway, I am still persuing this one, I will try some base resistors and an output network along with a closed loop control and it should make a decent amplifier. ^^

Dave47

Mon Oct 06 2008, 01:00AM

Registered Member #84 Joined: Thu Feb 09 2006, 01:06PM
Location: Dallas, TX
Posts: 47

Steve McConner wrote ...

Hint, Zobel network on the output and stopper resistors on your NTE87/88 bases will stop any parasitic oscillations you may get as you turn up the beans. If you really want a lot of RF power, metal-gate MOSFETs are hard to beat.

Also, you probably ordered the wrong op-amps. A 15MHz GBW does not mean it does anything worth a crap at 15MHz. My colleague wanted to build an amp with a gain of 10 at 44MHz, and he used an op-amp with a 1.8GHz GBW. It worked fine. I've used the AD811 to drive 50 ohm loads up to 40MHz. Two of them in parallel for extra power.

I like Zobel networks. I've had to use a few in the past.

I still like this op-amp...

Since it is a dual, you should be able to wire it up and get 40Vpp differential into a 25 ohm load at 10MHz (see figure 38).

David

Hazmatt_(The Underdog)

Mon Oct 06 2008, 02:54AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

That's a sweet amp but I'm not sure how to get 40Vpk pk out of +- 15V supply unless we're talking two supplies?

I have some THS4141's that I could use to drive both of those amp units with, and I might be able to really push them to high voltage output by adding their outputs. Might be fun... might be a lot of smoke... have to order a few and see. heh.

previous 1 2 3 4 next last

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