Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Instruments and Amps Everything that makes music, Especially including instrument amps.

Tube Emulation & EQ
Tube Emulation & EQ
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 7th February 2018, 06:39 PM   #341
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Markw4 View Post
Reminiscent of Bob Cordell's distortion magnifier
I had to look that up. You're absolutely right, exactly the same idea, but used for an entirely different purpose!

-Gnobuddy
  Reply With Quote
Old 7th February 2018, 06:51 PM   #342
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Printer2 View Post
I think getting 30V Fets to work would be a challenge enough.
If I understand KMGs circuit properly (which I might not), all the magic happens at the gate and source of his MOSFETs, via those additional diodes, resistors, and negative supply rail. Those components variously set DC bias, AC gain, and simulate grid current flow.

That means there is actually nothing special about the drain side of KMGs circuitry - just a regular drain resistor and positive supply voltage. As usual, the drain resistor sets the amount of voltage gain of the stage (in conjunction with the stuff at the FET source), and the B+ voltage sets the DC voltage at the drain.

If I'm correct about this, in principle, there should be nothing to prevent one from dialing down B+, and dialing down the drain resistor value at the same time to preserve reasonable biasing (drain at about half B+).

Voltage gain per stage will fall, of course. So will available output headroom. Those changes might potentially force other changes as well (such as needing one more gain stage). So lowered B+ is probably not the easy way to "clone" a classic valve preamp that was designed to run on 300V DC.

But, while not the easy route to a direct, one-for-one clone of a valve preamp, I think KMGs circuit will still emulate a triode even with reduced B+, the caveats being you have to make do with less gain, and less output headroom.

The big question is how far you can lower B+ before the lack of gain and output headroom become serious issues. Thirty volts is a long way down from three hundred! But I would think it wouldn't be too hard to make things work on, say, a B+ of 100 V instead of 300 V. Just a guess.

Fortunately, LND150s are cheap, and those 390V boost converters are cheap, and even a Triad N48-X 120:120 isolation transformer is only $14 USD if you prefer to use a more traditional solution (60 Hz transformer, voltage doubler).

Are you thinking about 30V mainly to make it safe for beginners to electronics? Or because of the convenience of running pre and (solid state) power amp off the same supply voltage?

-Gnobuddy
  Reply With Quote
Old 8th February 2018, 02:18 AM   #343
Printer2 is offline Printer2  Canada
diyAudio Member
 
Join Date: Apr 2010
Quote:
Originally Posted by Gnobuddy View Post
Are you thinking about 30V mainly to make it safe for beginners to electronics? Or because of the convenience of running pre and (solid state) power amp off the same supply voltage?

-Gnobuddy
Yeah I am thinking of the pedal community. For myself I am used to having high voltage around. I have a healthy respect for it at work feeding three phase 600V. The only time I was really nervous of 600V with 300A per leg was when I was troubleshooting a panel on an electric heat treating furnace. You could feel the power standing next to it. I just put on my clamp on ampmeters on the legs and took a couple of steps back. You feel real nervous twisting the pot with a screwdriver.

It would also be convenient if you could get it to work on low voltage. You could cook up a bunch of your favorite amplifier circuits and fit them in a pedal sized box. Strip board is fine up to a certain level of voltage, I remember splitting the cathode resistor on my mini 5E3 amp and needed a trace. When I put it together it worked but there was a ticking sound. I thought, 'great, I hate finding intermittent problems'. When I looked inside the chassis I heard the ticking and just to make it easy for me to find the problem the spot where I borrowed the trace from there was a little flash of light from the high voltage to the low voltage trace. The voltage arced over, built up, arced over...

LND150's are cheap, I just hate the shipping charge being $20 for $5 in parts.
  Reply With Quote
Old 8th February 2018, 06:17 PM   #344
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Printer2 View Post
I have a healthy respect for it at work feeding three phase 600V.
600 V scares the heck out of me.

I have a lifetime of experience with low-voltage electricity, but no formal training with high-voltage safety. So yeah, healthy respect mixed with a pang of fear, and wrapped in layers of all the good high-voltage safety information I have been able to teach myself.

Quote:
Originally Posted by Printer2 View Post
It would also be convenient if you could get it to work on low voltage. You could cook up a bunch of your favorite amplifier circuits and fit them in a pedal sized box.
I've been thinking about this a little bit.

Firstly, we can ignore saturation voltage. Today's FETs seem happy to work down to as little as 2 volts between drain and source, so saturation voltage is virtually a non-issue even with a low 30V B+. (Unlike a 12AX7, which is going to get increasingly unhappy if it has less than 100V between anode and cathode.)

With FET transconductance fixed, and saturation voltage negligible, the available voltage gain is proportional to the supply voltage (more precisely, to the DC voltage drop across the drain resistor).

So cutting B+ to 30V instead of 300V will drop the voltage gain of a stage to one-tenth as much. But one-tenth the voltage gain and one-tenth the signal amplitude still fit well together - peak to peak signal swing will be the same fraction of the B+ voltage as it was earlier!

So, at first glance, headroom will also be about the same as before. I wasn't expecting this, but I think it's true for KMG's circuit. All the desirable non-linearity happens because of Vgs, so Vds can be ten times smaller, and distortion will be the same as before, as long as Vgs is the same as before.

Since voltage gain is one-tenth as much, if we use the same number of gain stages, output voltage will, of course, be ten times smaller.

So what does have to change is the sensitivity of the power amp (signal needed for full output). It needs to be ten times more sensitive than the original all-valve power amp we're trying to clone.

But solid-stage gain is cheap and easy - it won't be too hard to make a power amp that only needs instrument-level (100 mV) input for full output. A single transistor stage set for a gain of 10x, in front of a typical class-D module with a 1V sensitivity, will do the trick.

Maybe your low B+ idea will work out better than we thought?

Quote:
Originally Posted by Printer2 View Post
The voltage arced over, built up, arced over...
You built a neon-bulb relaxation oscillator, without a neon-bulb!

Quote:
Originally Posted by Printer2 View Post
LND150's are cheap, I just hate the shipping charge being $20 for $5 in parts.
Too true. What can I say, the era of cheap transportation is over. I remember when gas in the USA was the equivalent of about 20 cents per litre Canadian...and that was three times the price the "old timers" I know talk about.

Back in 2018, if I drive 100 km round-trip to the "local" electronics store, my car sucks down about $13 worth of gas, not to mention other operating costs (oil changes, tyre wear, depreciation). The Canadian Revenue Agency suggests 50 cents/mile mileage allowance, so the bean-counters think that this 100km trip costs me about $50 all told.

That makes the $20 cost of shipping start to look good by comparison!

The only way I know to reduce the pain a bit is to queue up my purchases until I have $100 worth, and then shipping is free from either Digikey or Mouser. But on my present budget, it takes a long time to get to where it's worth buying $100 worth of electronics bits and pieces.

-Gnobuddy
  Reply With Quote
Old 8th February 2018, 06:43 PM   #345
Printer2 is offline Printer2  Canada
diyAudio Member
 
Join Date: Apr 2010
I'll have to go back and check but I thought it was a low voltage group effort but went high voltage using KMG's accomplishments. It would be interesting to see if they translate to low voltage in a convincing manor.
  Reply With Quote
Old 8th February 2018, 11:50 PM   #346
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Printer2 View Post
It would be interesting to see if they translate to low voltage in a convincing manner.
LTSpice seems to agree with my reasoning in my last post. See attached screenshots. (Note these are just LTSpice simulations right now - not built and tested and verified to work.)

I've changed the positive supply rail to only 30V. The drain resistor has been changed so the drain sits at about 15V DC quiescent. I used a 5V negative supply because 5V USB chargers are everywhere now and will work perfectly to provide the -5V supply rail.

Id is a bit under 1 mA.

KMG's stuff at the FET source has been slightly simplified - no Schottky diodes, just 1N4148s.

Small-signal gain is only about 3.5 times (11 dB). As predicted earlier, this will be the major compromise with drastically lowered supply voltage. But the low gain is necessary to maintain headroom at the lower B+, anyway.

The output waveform is shown for three input levels: 10 mV peak, 350 mV peak, 700 mV peak. It seems to look about right, but I don't know how it will sound until I listen to it.

The circuit will accept 1.5 V peak input without clipping (3 volts peak-to-peak.)

In the simulation, hard (and probably unpleasant-sounding) clipping occurs at 4V peak to peak input.

This seems plenty to cope with maximum guitar output under realistic conditions. (I'm still shocked at Michael Koster's 10V peaks, but I think most of us won't get past ten percent of that.)

There should probably be a clean gain stage ahead of the circuit shown here for use with an actual guitar. The circuit works as shown when fed from a low-impedance signal source, so a buffer will be required with an actual guitar, and I suspect some gain would be a good idea, too, if useful distortion is wanted without having to whale on the guitar strings with a sledge-hammer.

I should probably mention that clean transparent gain and buffering is trivially simple with today's solid-state components. We can probably get what we want out of a single JFET stage ahead of the KMG stage shown in this post (tweaking the 68k in the attached schematics to allow for the output resistance of the JFET.)

In fact, transparent-clean gain stages can be sprinkled through as necessary. KMG-stages to generate "tubeyness", clean solid-state stages to bring up the gain if necessary, without changing the sound otherwise.

-Gnobuddy
Attached Images
File Type: png KMG_30V_10mV_signal_2018-02-08.png (55.6 KB, 83 views)
File Type: png KMG_30V_350mV_signal_2018-02-08.png (182.0 KB, 84 views)
File Type: png KMG_30V_700mV_signal_2018-02-08.png (55.4 KB, 84 views)
  Reply With Quote
Old 9th February 2018, 01:15 AM   #347
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
We are discussing the use of fets and other parts to build a circuit that emulates a tube......why not just use a tube, or a tube and a fet?

There are subminiature tubes that were made for everything from hearing aids to missile guidance systems. I have been tinkering with some of them for two specific purposes, A preamp that can fit inside a guitar and run for hours from a small battery, and for modules in a Eurorack synthesizer. Both have space and power requirements in line with sand state design, but there are some tubes that should be useful, all are priced around $1 to $2:

The 6418 consumes little power. The DH filament draws 10 mA from 1.25 volts. Plate voltage? 30 volts MAXIMUM, 15 to 22.5 recommended. Power output, 2 or 3 milliwatts. The issue, microphonics. Not such a big deal if mounted inside the guitar. It adds some "air" to the sound. Not for use in a combo amp!

The 6088 consumes a bit more power, and is far less microphonic since it has a thicker filament. It needs 20 mA at 1.25V. Max plate voltage, 67.5 although I have tested it to 125 volts without issue. Recommended B+ voltage, 45 volts, but it works good on 25 volts and still amplifies on 9 volts. Power output, a big 10 milliwatts!

The 5678 is the big boy here. Its filament needs 50 mA, and it's plate is rated for 90 volts, but I still didn't go beyond 125 V. It was made for RF use, hence no power output spec.

I'm looking at these for audio amplifier use, and all 3 make good triodes. The triode curves are in the data sheets.

You can operate them as a pentode with a mosfet load / buffer like I did in my HBAC Amp 1.5. The gain can be turned way up for some super saturated sounds, but microphonics become an issue with any tube when the stage gain gets into the hundreds.

There are all flea powered directly heated tubes. There are also a whole bunch of indirectly heated tubes that have 6.3 volt heaters, but they are beyond my power budget (I need a tube per string for piezo saddles), and probably that of a battery powered pedal.
__________________
Tubelab, it's 5 year mission. To explore strange new tubes, to seek out new circuits and topologies, to boldly go where no tube has gone before......
  Reply With Quote
Old 9th February 2018, 01:17 AM   #348
Printer2 is offline Printer2  Canada
diyAudio Member
 
Join Date: Apr 2010
Your going to hate me. So what happens if you swap in a 2N5458 or the like?

I almost ordered a few pencil tubes. That might be next after I try some SS.
  Reply With Quote
Old 11th February 2018, 05:41 AM   #349
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Tubelab_com View Post
...A preamp that can fit inside a guitar and run for hours from a small battery
I have one Takamine acoustic guitar that came with the manufacturer's "Cool Tube" piezo preamp. It uses a 12AU7, runs off four AA cells, and a fresh set of cells lasts for an incredibly long time. I bought the guitar in 2014, and didn't replace the original Japanese-made AA cells until the end of 2017 (though admittedly, this guitar didn't get a lot of plugged-in playing time for much of that period.)

I don't have a schematic for the Cool Tube preamp, so I don't know what Takamine actually designed the tube to do. The "cool tube" name is supposedly to reassure you that there isn't a hot vacuum tube baking the moisture out of your precious guitar's wooden body. The insanely long battery life suggests there isn't a whole lot of power going to the heater. And I don't know if there is any internal boost converter, or the 12AU7 only has 6V on its anode.

What I do know is that this particular guitar has less of that harsh piezo sound (some call it "piezo quack") than most piezo-equipped guitars I've heard. Whether the 12AU7 is actually responsible for that or not, I don't know.

-Gnobuddy
  Reply With Quote
Old 11th February 2018, 06:04 AM   #350
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Tubelab_com View Post
There are also a whole bunch of indirectly heated tubes that have 6.3 volt heaters
I bought a few of those back around 2010 or so, when I had just made the earthshaking discovery that the awful electric guitar tone I'd had for 25 years, had been cured 30 seconds after plugging into my first valve guitar amp.

I put them away not long after I bought them, once I realized I really didn't have a clue how a valve guitar amp worked its magic (I knew something about solid-state Hi-Fi audio electronics, sure, but those sounded awful for guitar...)

It's been a while since I looked at those little subminiature tubes, but I think I have a couple of triodes, a small-signal pentode, and a couple of 5902 4-watt power "pentodes" (which are probably actually beam tetrodes.)

Do we know if the guts of any of the indirectly heated subminis came from pre-existing larger valves? It would be funny if it turned out a 5902 is just a repackaged 6AS5 or something similar (with tweaks to make it resist high G forces and what not).

-Gnobuddy
  Reply With Quote

Reply


Tube Emulation & EQHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
CA 3080/LM 3886 tube pwr amp emulation metalhead Solid State 2 3rd February 2018 03:10 PM
SIT Device Emulation poynt99 Pass Labs 1 27th April 2014 09:14 AM
Tube emulation with transistors fireworks Tubes / Valves 17 4th February 2014 06:03 PM
Tube emulation (pritchard article) Barre Instruments and Amps 7 4th November 2010 01:28 PM
Tube amp semiconductor (FET?) emulation danV Instruments and Amps 11 24th June 2008 09:40 PM


New To Site? Need Help?

All times are GMT. The time now is 02:15 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki