ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Re:Minibooster Operation

9/8/1999 10:23 PM
R.G.	Re:Minibooster Operation I'll have to pull out some old texts to be sure, but I think this is what is going on. The two 1M's tie the gate of the first JFET at 1/2 the supply voltage. If the source were at exactly the same voltage as the gate, the current through the device would lie on the Vgs=0 line, which means that over the pinchoff voltage, the current would be Idss, from a few to maybe 20ma, depending on the JFET. The bottom JFET is held at Vgs=0, so it's trying to pull its own Idss through the top device. This pulls down on the top JFET's source turning it more on until the top device hits Idss also. With both devices trying to pull Idss, they will come to some kind of balance, with the voltage at the source of the top and drain of the bottom device determined by the relative matching of the two devices. If they're perfectly matched, the voltage at the output will be 4.5V, the same as the gate voltage on the top device. The gate and source of the top device are held together by that 0.01uf cap, so there is no AC voltage between the two. That means that the top JFET will act like a constant current source. The bottom JFET is biased at DC by the interaction of the two, and it sees a constant current source drain load, and a zero source load, so it runs at the maximum possible gain for the device. I simmed the thing and it looks like a gain of 20 to 50 depending on the device. This is an almost exact dupe of a triode circuit from a few decades ago, and works much the same way. Yeah, if you get some high Idss devices, it could pull a lot of current. Idss is commonly 5-15 ma for the recommended JFETs; a 9V battery is a 165ma-hr device, so you could get as little as 10 hours out of a battery, and it would run down quickly. As the transistors run the battery down, they will be balanced on the edge of the JFET triode region, so the tone will change all right.

9/8/1999 10:23 PM

R.G.

Re:Minibooster Operation

I'll have to pull out some old texts to be sure, but I think this is what is going on.

The two 1M's tie the gate of the first JFET at 1/2 the supply voltage. If the source were at exactly the same voltage as the gate, the current through the device would lie on the Vgs=0 line, which means that over the pinchoff voltage, the current would be Idss, from a few to maybe 20ma, depending on the JFET.

The bottom JFET *is* held at Vgs=0, so it's trying to pull its own Idss through the top device. This pulls down on the top JFET's source turning it more on until the *top* device hits Idss also.

With both devices trying to pull Idss, they will come to some kind of balance, with the voltage at the source of the top and drain of the bottom device determined by the relative matching of the two devices. If they're perfectly matched, the voltage at the output will be 4.5V, the same as the gate voltage on the top device.

The gate and source of the top device are held together by that 0.01uf cap, so there is no AC voltage between the two. That means that the top JFET will act like a constant current source. The bottom JFET is biased at DC by the interaction of the two, and it sees a constant current source drain load, and a zero source load, so it runs at the maximum possible gain for the device. I simmed the thing and it looks like a gain of 20 to 50 depending on the device.

This is an almost exact dupe of a triode circuit from a few decades ago, and works much the same way.

Yeah, if you get some high Idss devices, it could pull a lot of current. Idss is commonly 5-15 ma for the recommended JFETs; a 9V battery is a 165ma-hr device, so you could get as little as 10 hours out of a battery, and it would run down quickly. As the transistors run the battery down, they will be balanced on the edge of the JFET triode region, so the tone will change all right.

9/9/1999 1:36 AM
Joe Fuzz	R.G.: Thanks for the analysis. Just so I'm perfectly clear on this: you're saying that the circuit is ON all of the time? In other words, there is a constant DC current draw even with the AC input at GND and that this DC current could be as high as 15 mA. Correct? If so, I think that answers my next question which was "would a higher value cap across the battery help?" If there was a current surge when the input went high then I could see a bigger cap helping. However, what we seem to be talking about here is a steady drain on the battery even with no AC signal. BTW, when I first built this box last year, I ran a test on the DC voltage at the point where Q1's Source and Q2's Drain meet. Since these are JFETs, I was concerned about the non-linearity of the devices. I varied V_DD from 10V down to 4V. The voltage at that "halfway" point stayed at 1/2 of V_DD throughout the range (give or take 30mV or so). It seemed linear. That's why I was so puzzled when the battery slipped only half a volt and the sound changed. (I use Duracells which seem to be ~9.55V out of the box and the battery I removed was only down to ~8.9V.) Now that I think about it, maybe I was confusing apples and oranges. Maybe I was slipping down the JFET curve -- in a "linear" fashion. If that makes any sense...

9/9/1999 1:36 AM

Joe Fuzz

R.G.: Thanks for the analysis. Just so I'm perfectly clear on this: you're saying that the circuit is ON all of the time? In other words, there is a constant DC current draw even with the AC input at GND and that this DC current could be as high as 15 mA. Correct?

If so, I think that answers my next question which was "would a higher value cap across the battery help?" If there was a current surge when the input went high then I could see a bigger cap helping. However, what we seem to be talking about here is a steady drain on the battery even with no AC signal.

BTW, when I first built this box last year, I ran a test on the DC voltage at the point where Q1's Source and Q2's Drain meet. Since these are JFETs, I was concerned about the non-linearity of the devices. I varied V_DD from 10V down to 4V. The voltage at that "halfway" point stayed at 1/2 of V_DD throughout the range (give or take 30mV or so). It *seemed* linear. That's why I was so puzzled when the battery slipped only half a volt and the sound changed. (I use Duracells which seem to be ~9.55V out of the box and the battery I removed was only down to ~8.9V.)

Now that I think about it, maybe I was confusing apples and oranges. Maybe I was slipping down the JFET curve -- in a "linear" fashion. If that makes any sense... :-)

9/9/1999 3:06 AM

R.G.

quote:
" Just so I'm perfectly clear on this: you're saying that the circuit is ON all of the time? In other words, there is a constant DC current draw even with the AC input at GND and that this DC current could be as high as 15 mA. Correct?"

Yep. As long as you put a voltage supply on the drain of the top JFET, this thing will run at close to Idss for the JFET. Some JFETs have low Idss, in the units of milliamps, but some are fairly high. Depends entirely on the type number and specific device.

It's *not* a push-pull circuit, even thought it might look like it. It's a linear, class A amplifier. The top JFET is just an active current source.

quote:
"Maybe I was slipping down the JFET curve -- in a "linear" fashion. If that makes any sense."

It makes a lot of sense - that's exactly what was happening. The two devices stayed balanced, but were moving into their triode region, I'd guess, so the tone changed.

You could go find low Vgs-off JFETs. These things have a lower Vth and so they would stay in channel saturation (hence out of triode mode) to lower voltages than higher Vgs types. 2N5485 would be a good place to start.

9/9/1999 5:51 AM

Joe Fuzz

quote:
"The two devices stayed balanced, but were moving into their triode region, I'd guess, so the tone changed."

I guess I could quantify what I mean by "tone change." The circuit does soft clipping quite well. Excellent, in fact as in 12AX7 excellent. But it doesn't do hard clipping very well -- at least, I couldn't ever get it to sound good when overdriven. Did you ever cascade SHOs? That's what it sounded like. I'm frankly mystified by the "cascaded mini-booster" up at the PCB site. I haven't actually inspected the circuit but I cascaded 2 mini-boosters and it sounded awful to me.

What is happening now is I'm getting that same brittle cascaded tone but with only one stage. I'm *guessing* that I have slipped up or down the load line and that I'm now banging one of the rails harder than I normally would.

quote:
"2N5485 would be a good place to start."

Thanks for the suggestion.

9/9/1999 5:57 PM
Aron	Exactly, it doesn't do hard clipping very well - but it does a soft to medium clipping very well into a good tube amp. Now, I DO have a version of the cascaded MiniBooster - my Booster+ (3 of them with Radio Shack MPF102's) that does do harder clipping well - almost like a Big Muff. It has something to do with the gain of the booster. That Booster+ had unusually low gain. The J201's have very high gain. Yes, I have cascaded SHO's and 2 MiniBoosters with the different combinations I list on my site do not sound anything like the cascaded SHO's to me. Aron

9/9/1999 5:57 PM

Aron

Exactly, it doesn't do hard clipping very well - but it does a soft to medium clipping very well into a good tube amp.

Now, I DO have a version of the cascaded MiniBooster - my Booster+ (3 of them with Radio Shack MPF102's) that does do harder clipping well - almost like a Big Muff.

It has something to do with the gain of the booster. That Booster+ had unusually low gain. The J201's have very high gain.

Yes, I have cascaded SHO's and 2 MiniBoosters with the different combinations I list on my site do not sound anything like the cascaded SHO's to me.

Aron

9/9/1999 6:03 PM
R.G.	Say, Aron - you get those MB2 boards yet?

9/9/1999 6:09 PM
Aron	Hey, Yikes I forgot to write back. They are very nice! I populated one with a few mods and will be trying it out soon. Very nice!

Page 2 of 3