I am inclined to think so, based on playing the Bloc 100G clean channel for a few decades (!). Looking at single Jfet stages, and Jfet/BJT pairs gives a few options as to where these stages can be placed in the preamplifier section. The nice thing about Jfet/BJT pair compared to a BJT/BJT I think is the advantage the input impedance can be kept high and still having predictable and stable voltage gain at the output. What also may be of interest is whether the JFET/BJT pair gets some "bias shifting" if it is hit with a large input signal transient.
Here we have the all-FET Randall Guitar Preamp from their famous RG100
Funny thing is that clipping on the Dirty channel is made from back to back Zeners or Diodes!!!!, which is an oxymoron 😱
FWIW distortion is buzzy, the trademark sound of Thrash Guitar players, notably Pantera´s Dimebag Darrel, who famously said "I had been using Tube amps, but was instantly fascinated by its buzzsaw sound" 😱
It also uses a constant current Power Amp which enhances the effect.
Funny thing is that clipping on the Dirty channel is made from back to back Zeners or Diodes!!!!, which is an oxymoron 😱
FWIW distortion is buzzy, the trademark sound of Thrash Guitar players, notably Pantera´s Dimebag Darrel, who famously said "I had been using Tube amps, but was instantly fascinated by its buzzsaw sound" 😱
It also uses a constant current Power Amp which enhances the effect.
If actual Zeners were used, that'd be less so, AFAICR. You'd still need a blocking diode in series - and a lot more voltage to drive it than this simple 1N914 arrangement. I admire how they didnt deviate from the classic fender style tone "stack" lest no one buys the amp due to those controls working differently.
As some of you know, I recently developed an LTSpice circuit that delivers a simulated guitar signal, closely matching one of the real guitar captures on Rod Elliott's website. Details are in another thread ("Unwanted clipping in solid-state e-guitar preamps", here: https://www.diyaudio.com/community/...pping-in-solid-state-e-guitar-preamps.391513/ )
The reason I developed that simulated guitar signal, and posted details and a working LTSpice simulation, was to allow all of us to get better, more realistic results from LTSpice simulations of guitar amplifiers and preamplifiers. Instead of testing with a simulated sine wave, we can test with a signal that reveals how the circuit will behave with an actual guitar plugged in.
Here's an example. The attached screenshot shows the result of connecting that simulated guitar signal to a simulated JFET-BJT compound stage (straight from the Traynor Bloc 100G schematic that Shanx posted earlier in this thread).
I've scaled the red trace (using R5/R6) so that it exactly overlays the blue trace if you feed the JFET/BJT pair a very small signal, such as 1mV peak. As the signal amplitude gets better, the JFET "squashes" the signal, so for big input signals, the output no longer exactly overlays the input.
Overlaying the red and blue traces allow us to see exactly what's happening in the JFET/BJT compound gain stage.
As we suspected, we can now see that the output signal from the JFET-BJT pair (red trace) is slightly squashed compared to the input signal (blue trace), for about 500 mS at the start of the note.
While running the simulation, I zoomed into the graph, and you can see that the negative peaks of the red signal show no signs of hard clipping at all. In fact I can't really see any change in the shape of the waveform peaks when comparing the red and blue traces. The only obvious difference is that the peaks of the red signal have been gently squashed, i.e. amplitude is slightly reduced.
This is the effect of only one single JFET-BJT stage. In a guitar amplifier, the signal would go through several such stages, and each one would add a little more gentle peak squashing.
This (Traynor Bloc 100G) is also just about the first time I've seen an intelligently designed JFET bias scheme used in a guitar amplifier. The use of a negative supply rail, along with negative DC feedback from the BJT collector, makes the circuit forgiving of JFET parameter variations.
Back in the early 1980s, I came up with almost this exact idea in my very first JFET guitar onboard preamp. It used a single supply rather than +/- 15V, but it used the same type of DC feedback from the collector of the PNP transistor to the source of the N-channel JFET.
I haven't seen this circuit used elsewhere until now.
Unlike most of the JFET guitar circuits on the 'Web, this one (Bloc 100G) will actually bias itself reasonably, and work properly, if you plug in any reasonable JFET.
The Randall R100 schematic Fahey posted shows the more typical approach. The circuit will only work if you use JFETS selected to have a very narrow spread of characteristics. A big manufacturer like Randall might have been able to order pre-sorted JFETs straight from the manufacturer. But a DIY electronics enthusiast using store-bought JFETs would probably find that the circuit would not work properly.
All in all, I definitely give the JFET-BJT compound stage a thumbs-up. It's definitely worth investigating more closely.
The tragedy is that there are always far more interesting ideas to pursue, than there is free time to actually pursue them. 😀
Currently most of my free time is going into building a little P.A. powered speaker for use at our weekly music jams.
-Gnobuddy
The reason I developed that simulated guitar signal, and posted details and a working LTSpice simulation, was to allow all of us to get better, more realistic results from LTSpice simulations of guitar amplifiers and preamplifiers. Instead of testing with a simulated sine wave, we can test with a signal that reveals how the circuit will behave with an actual guitar plugged in.
Here's an example. The attached screenshot shows the result of connecting that simulated guitar signal to a simulated JFET-BJT compound stage (straight from the Traynor Bloc 100G schematic that Shanx posted earlier in this thread).
I've scaled the red trace (using R5/R6) so that it exactly overlays the blue trace if you feed the JFET/BJT pair a very small signal, such as 1mV peak. As the signal amplitude gets better, the JFET "squashes" the signal, so for big input signals, the output no longer exactly overlays the input.
Overlaying the red and blue traces allow us to see exactly what's happening in the JFET/BJT compound gain stage.
As we suspected, we can now see that the output signal from the JFET-BJT pair (red trace) is slightly squashed compared to the input signal (blue trace), for about 500 mS at the start of the note.
While running the simulation, I zoomed into the graph, and you can see that the negative peaks of the red signal show no signs of hard clipping at all. In fact I can't really see any change in the shape of the waveform peaks when comparing the red and blue traces. The only obvious difference is that the peaks of the red signal have been gently squashed, i.e. amplitude is slightly reduced.
This is the effect of only one single JFET-BJT stage. In a guitar amplifier, the signal would go through several such stages, and each one would add a little more gentle peak squashing.
This (Traynor Bloc 100G) is also just about the first time I've seen an intelligently designed JFET bias scheme used in a guitar amplifier. The use of a negative supply rail, along with negative DC feedback from the BJT collector, makes the circuit forgiving of JFET parameter variations.
Back in the early 1980s, I came up with almost this exact idea in my very first JFET guitar onboard preamp. It used a single supply rather than +/- 15V, but it used the same type of DC feedback from the collector of the PNP transistor to the source of the N-channel JFET.
I haven't seen this circuit used elsewhere until now.
Unlike most of the JFET guitar circuits on the 'Web, this one (Bloc 100G) will actually bias itself reasonably, and work properly, if you plug in any reasonable JFET.
The Randall R100 schematic Fahey posted shows the more typical approach. The circuit will only work if you use JFETS selected to have a very narrow spread of characteristics. A big manufacturer like Randall might have been able to order pre-sorted JFETs straight from the manufacturer. But a DIY electronics enthusiast using store-bought JFETs would probably find that the circuit would not work properly.
All in all, I definitely give the JFET-BJT compound stage a thumbs-up. It's definitely worth investigating more closely.
The tragedy is that there are always far more interesting ideas to pursue, than there is free time to actually pursue them. 😀
Currently most of my free time is going into building a little P.A. powered speaker for use at our weekly music jams.
-Gnobuddy
Attachments
Last edited:
Thanks for posting up the simulation of it. The Bloc 100G was designed in the early 80's and I think Traynor did a really good job. Of course for a lot of solid state amps that have been manufactured in the decades since, there are relatively few discrete, amps without op-amps. It`s nice to explore some options with regard to design of electric guitar and bass preamps.
Traynor/Yorkville has a good reputation for reliability and quality here in Canada. Most musicians I've met here either own, or have owned, Traynor gear.
I moved to BC from Los Angeles some years ago. I had never heard of Traynor during all my years in Los Angeles. That's just one more example of the wide cultural gulf that exists between the USA and Canada, even though the two countries are physically adjacent and have a long shared border.
Wikipedia has some interesting information about Traynor: https://en.wikipedia.org/wiki/Traynor_Amplifiers
At a jam this summer, one musician showed up with one of these: https://yorkville.com/loudspeakers/excursion/product/exm-mobile/
This is something I've been wanting for years - a battery operated, compact P.A. speaker with really good audio quality, capable of enough SPL for use at a typical small jam, using good-quality drivers. No nasty-sounding piezo horn tweeters, no nasty-sounding compression drivers. Compromised on maximum SPL capability, rather than on sound quality.
I liked the sound quality of the little Yorkville immediately, but not the price: https://www.long-mcquade.com/97309/...e-Sound/Excursion-Mini-Battery-Powered-PA.htm
I've been trying to "roll my own" version of something rather similar for a while. My first attempt used a pair of thrift-store-refugee Sony loudspeakers from a long-ago Sony "mini component Hi-Fi" system. Unfortunately audio quality of the Sony speakers turned out to be sub-par, and I'm now trying to build something better.
But that's another story altogether, for another thread, once I get the thing finished!
-Gnobuddy
I moved to BC from Los Angeles some years ago. I had never heard of Traynor during all my years in Los Angeles. That's just one more example of the wide cultural gulf that exists between the USA and Canada, even though the two countries are physically adjacent and have a long shared border.
Wikipedia has some interesting information about Traynor: https://en.wikipedia.org/wiki/Traynor_Amplifiers
At a jam this summer, one musician showed up with one of these: https://yorkville.com/loudspeakers/excursion/product/exm-mobile/
This is something I've been wanting for years - a battery operated, compact P.A. speaker with really good audio quality, capable of enough SPL for use at a typical small jam, using good-quality drivers. No nasty-sounding piezo horn tweeters, no nasty-sounding compression drivers. Compromised on maximum SPL capability, rather than on sound quality.
I liked the sound quality of the little Yorkville immediately, but not the price: https://www.long-mcquade.com/97309/...e-Sound/Excursion-Mini-Battery-Powered-PA.htm
I've been trying to "roll my own" version of something rather similar for a while. My first attempt used a pair of thrift-store-refugee Sony loudspeakers from a long-ago Sony "mini component Hi-Fi" system. Unfortunately audio quality of the Sony speakers turned out to be sub-par, and I'm now trying to build something better.
But that's another story altogether, for another thread, once I get the thing finished!
-Gnobuddy
Attachments
I stumbled into a Ion "Blockrocker" earlier this fall for next to nothing and have resurrected the 12V lead acid battery in it (I hope) via rejuvenation via flyback pulses into it. The little portable amp doesnt sound well, which I attribute to the speakers and analog processing done by Ion in the design. It's got some TI class D in it, which should do for the acoustic music I play. I opened it up, but they placed the MB component side down so I didnt go in further; just put it back together for another time.
Another tack I'm considering is the "W,FIT" one and just get a pure sine inverter, which are cheap enough these days. 300W should do and I've read the quiescent current isnt too bad, for low load use, such as powering a wal-wart based mixer, compressor, and some small 20W AC powered amp with digital effects I already have.
Something like this https://www.ebay.com/itm/333275322180
Last edited:
Here is a snippet of the Bloc 100G "overdrive" channel. It's a JFet-BJT-BJT combo and in the local NFB you have a pair of standard clipping diodes D3&D4. As I mentioned before the overdrive channel sounds alright, nothing really outstanding as far as distortion goes it's pretty mild and basic. I might get some time to play around with this part of the circuit and improve on it. The Clean channel sounds great as is, and I have used that with external pedals for drive. My jfet booster with soft limiting going into the Clean channel definitely sounds more "gling". The soft limiter is single diode with series and parallel resistor in the local feedback, so it gives a nice asymmetric squish on one side of the waveform.
Reminds me of the 80's where I heard these speakers, maybe the wood tamed the compression drivers? One of the better speakers I ever heard.
Reminds me of the 80's where I heard these speakers, maybe the wood tamed the compression drivers? One of the better speakers I ever heard.
Argh. Had a weird glitch either between my ears, or between Firefox and the forum software; net effect, and empty post that was hidden from me for 54 minutes, so now I can't delete or edit it.
If a moderator reads this - could you kindly delete my post #70, just above this?
Thanks!
-Gnobuddy
If a moderator reads this - could you kindly delete my post #70, just above this?
Thanks!
-Gnobuddy
I have heard good-sounding compression tweeters on small P.A. systems myself - but only twice that I can remember.
The first time I was sitting outdoors on Catalina Island (offshore from Los Angeles) near a restaurant, waiting for the ferry to arrive. There was a band playing live music for the outdoor seating area of the restaurant. They had fairly small black P.A. speakers pole-mounted high enough for me to see them over the low fence. I was close enough to hear that the sound quality was excellent. Not having any way to make a note, I memorized the brand of the P.A. speakers...only to forget before I made it back home several hours later.
Embarrassing, but I probably wouldn't have been willing to pay the purchase price even if I'd remembered. I don't think I was even active in music at all at that time - and even when I got back into it, I don't make any money playing music, I do it for love. That means I avoid expensive music gear.
The other time I heard good sound from a P.A. system with horn midrange/tweeters was at a family function in California for a relative's 75th birthday. A guitarist/cellist/vocalist duo who go by "Acoustic Eidolon" performed some music for the guests. They also had a pair of relatively small, pole-mounted, powered P.A. speakers.
Those speakers did use a compression driver / horn tweeter, but they sounded excellent. I looked up the price; it was a well-known brand, and all I can remember is that the pair cost far north of $1000 USD apiece at the time (some fifteen years ago). Way too much to hold any interest for me - so I unchecked that box on my mental list, and have since forgotten the brand and model.
As a result of those two experiences, I believe that it IS possible to make a P.A. system with compression drivers coupled to horns that does not sound horrid. But that's not the norm, it's the exception. And it comes at a high price.
To me, every affordable P.A. system I've seen (and many high priced ones) sound nasty in the midrange and treble. The common element is that they all use very similar looking horns, probably with very similar moving-coil compression drivers mounted behind them. The ones that use piezo horns sound even worse.
I'm definitely not the only one who doesn't like the sound of horn-loaded speakers. On Rod Elliott's website ( https://sound-au.com/articles/waveguides1.htm ) there is the following interesting bit of text (part of an article on horns from contributor Robert C White):
The usual paradigm in affordable P.A. speaker design seems to be "Being loud is more important than sounding good". It seems to work well for manufacturers, as the typical customer for those products agrees with that philosophy.
As for me, I don't have a large audience to satisfy, I don't play with a drummer, and I don't like high SPL levels (why do so many musicians seem bent on destroying their own hearing?) I don't need high SPL, and I can't stand the abysmal sound quality of a typical P.A. system.
That leaves me somewhat out in the cold. The conventional P.A. systems I can afford, I don't like. The ones I like, I can't afford.
The best solution I've found so far is my old Acoustic AG30, a wedge-shaped, bass-reflex enclosure holding a 8" mid-woofer with a post-mounted coaxial plastic dome tweeter, very like a larger version of the speakers used in good quality automotive audio systems (and also in high-end ceiling speakers used for background music). It's more than loud enough for me, and audio quality is far better than affordable small P.A.s with horn midrange/tweeters.
The only catch with the AG30 is its weight, bulk, and awkward shape. In 2010 I coped with it easily. Twelve years later, I have a few of the typical physical challenges that come with age, including a bout with bursitis (apparently a long-lasting side-effect of the COVID virus). It's getting harder to haul the AG30 up and down multiple flights of stairs, and in and out of the minivan.
All of which got me thinking about designing and building my own, even smaller personal P.A. system, with the focus on high sound quality, along with small weight and bulk. The plan is to compromise on maximum loudness, not on audio quality.
I think this is also what the designers of that little Yorkville P.A. aimed for.
-Gnobuddy
Sounds to me like what you want is the old Morel MDT37 tweeter. I used them in an absolutely-stupid-loud car audio rig that Ive since dismantled. Pro audio mids, and keeping up with twin LAB12’s. What killed them was heat from the sun - the adhesives holding the magnet softened and the magnet shifted, pinning the VC. After about 6 years. Not the best tweeter I’ve ever heard, but much more of the typical soft-dome sound since it’s just a shallow waveguide. I’m partial to true ribbons for SQ.
They may still make a version of it, I don’t know. Probably $200 apiece by now.
They may still make a version of it, I don’t know. Probably $200 apiece by now.
Almost bought one of those the other day! My luck coincided instead with a Behringer Ultracoustic ACX450. Appears to be nicely done, with separate EQ and digital effect for each input, instrument and mic. I hope it sounds good and is something I can handle. I once owned a Fender Twin. Lift that by its handle and you know what roadies are for.
Oh, speaking of "separate EQ and digital effect for each input" I had my eye on a Crate amp designed for the acoustic musician, which from the look of it, was actually two systems in one box; two 30W amplifiers and two speakers, separate channels; one for the instrument, one for the vocal. I've no idea how heavy that was, but I can find out. Seems to be a good way to go architecture wise.
Neither amp has built in compressors. My little Behringer QX1002 does - on 2 channels. One would think that would be a standout feature.
Last edited:
The Acoustic AG30 is actually lacking that - EQ and FX are shared by both inputs. This is the thing I like least about this amplifier, in fact. For example, the eq curve that suits my wife's voice best, is very sub-optimal for my own vocals.
But it isn't "boxy", and it isn't nasal, and it isn't harsh; and that makes its sound head and shoulders better than most similar gadgets I've tried.
There was briefly an Acoustic AG60, which appeared to be nothing more than two side-by-side AG30s built into one enormously heavy and incredibly cumbersome MDF enclosure.
It sounded great in the store, but I didn't own a tame pigmy elephant or a small forklift truck, so it had zero appeal.
I used to use a separate small Mackie mixer along with the AG30. More cabling to run, but the Mackie allowed me to dial in the best-sounding EQ curve for each of the three or four different singers using it, using one shared reverb setting on the AG30.
-Gnobuddy
I have a little good news, some places still teach this. This guy (Aaron Lanterman, I actually met him a couple times) also sometimes references the late Marshall Leach (Ga. Tech prof known for his audio designs), whose notes and such are still online. Lanterman also has a course on music synthesizers with youtube videos. He had already been recording some of his in-person lectures many years ago, but of course Covid accelerated him putting more online.
Thanks for that!
I hate to see an entire field of human knowledge slowly fade away to obscurity, and from that point of view, it's nice to know there are some attempts to keep this sort of thing alive.
On the other side of the coin, it's an interesting question, I think, whether it is a good thing to teach analogue discrete audio design in college engineering courses today, or not.
I think the field (discrete analogue design) is still necessary for some advanced applications, and as we've seen, it's also a good thing to know for use with electric guitars.
But those are both fairly obscure corner cases. For the vast majority of young students working towards their EE degree, their future will mostly deal with computers, software, and maybe some digital hardware.
If they have to deal with analogue stuff at all, it will most likely be limited to portions of a mixed-signal circuit (mixed digital and analogue), and maybe some front end circuitry for an ADC, which will probably be an IC or two, maybe a programmable gain amplifier or a couple of op-amps.
As for e-guitar amplifiers, their future is almost certainly going to be dominated by digitally modeled simulations of decades-old vacuum tube amplifiers, running on ever faster and ever more powerful DSP chips, or just running as a software DAW plugin on a host PC.
Though I don't like it, I think the honest truth is that discrete analogue audio design is a fading backwater of electronics, and it would be essentially a waste of time and money for the vast majority of EE students if they were forced to sit down and put hours of study into it.
-Gnobuddy
Not sure that I agree that there's little application for it. In addition to the obvious applications, it's a basis for getting into RF design, both for discrete and integrated applications (someone has to design the chips). I don't know how much of the higher frequency RF is BJT vs. FET type designs (Lanterman also has videos covering FETs), but there's certainly a range of RF where BJTs are useful and knowing how to bias them and such is needed, even if it's Class C.
Agreed, and I consider all those to be included in the category of "advanced applications" that I mentioned in post #77.
There are other oddball applications using discrete electronic components out there, like multi-kilowatt electronics control boards used with wind turbines. A former student showed me a PCB he was laying out for a company that made boards used in wind turbines - it included an IGBT nearly the size of a matchbox. The IGBT was rated for something like 1200 volts maximum voltage, 100 amps maximum current.
But I suspect most EEs graduating these days won't end up doing these relatively exotic things.
-Gnobuddy
Not sure about the Acoustic brand one ,if it´s from then 70s 80s, they DID use FET compressor/limiters but only in the big Bass Amps
Now if it´s from the resurrected Company, same with the Behringers they most probably do, simply there is no way to shut them OFF ... for good reason I might add.