Go Back   Home > Forums > >

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

The Hundred-Buck Amp Challenge
The Hundred-Buck Amp Challenge
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
Old 22nd November 2021, 07:38 AM   #2191
Thomas Schwarz is offline Thomas Schwarz  Germany
diyAudio Member
 
Thomas Schwarz's Avatar
 
Join Date: Nov 2020
Hi Gnobuddy,

First of all, thank you so much for your very thorough answer!

Quote:
Originally Posted by Gnobuddy View Post
...
It's been a few years, but IIRC, I had a pentode supply voltage in the vicinity of 150 volts, and the actual screen voltage was, let's say, 75 volts (it might have been less, I forget the exact value I ended up using.)

Let's say we decided to use two 100k resistors in series to make a voltage divider, providing 75 volts at the junction when unloaded.

However, the Thevenin (source) impedance of the voltage divider is 50k (the parallel resistance of two 100k resistors). If the screen current of the small signal pentode changes by 1 mA, the voltage at the junction drops by 50 volts! That is a lot of sag, especially if the initial voltage was only 75 volts!
...
After reading your comments I looked up the EF80's datasheet - which I clearly should have done before posting my previous remark. I was very surprised to find that this pentode istn't actually so "small signal" at all and draws 3+ mA at 75V Ug2 and Ug1 = 0V (https://frank.pocnet.net/sheets/084/e/EF80.pdf) ! I had something like an EF86 preamp first stage in mind (0.1 mA Ig2).

Quote:
Originally Posted by Gnobuddy View Post
...
I have a little experience building and testing out a few small-signal pentode gain stages in guitar amps, and I found that it was definitely possible to have a screen decoupling cap that was too big. This produces a very spongy, compressed sounding guitar amp that responds very slowly to the guitar, like a low-quality compressor set with a too-long decay time. It might be suitable for some types of music (slow blues solos, say), but is very poorly suited to fast guitar playing
...
So a power FET with the voltage fixed from an array of zeners (amplified zener diode) seems to be the way to go - maybe even if a suitable lower voltage PT tap were actually available. A stiff variable screen supply looks like a very attractive way to power-scale the amp ...

Cheers,
Tom
  Reply With Quote
Old 22nd November 2021, 07:48 AM   #2192
PFL200 is offline PFL200  Netherlands
diyAudio Member
 
Join Date: Jul 2021
Quote:
Originally Posted by Thomas Schwarz View Post
Just in case I'm not the only one that didn't think of using the CT for low-voltage screen supply - I've found this schematic that demonstrates exactly that - incl. the fact that a separate rectifier for the screen isn't required...
Attached the complete article and errata/notes on this amplifier. A diode across the 100 uF screen supply capacitor was recommended to prevent a negative voltage developing over that capacitor when the amplifier is in stand-by mode.

Philips used the CT for screen supply in some of their PA amplifiers, like the EL6426.
Attached Images
File Type: gif Errata and Notes.gif (9.0 KB, 56 views)
Attached Files
File Type: pdf Article.pdf (1.54 MB, 7 views)
File Type: pdf EF86 + ECC85 + 4 x EL36 pp (Philips EL6426).pdf (636.3 KB, 13 views)
  Reply With Quote
Old 22nd November 2021, 12:55 PM   #2193
Tubelab_com is online now Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
Originally Posted by Thomas Schwarz View Post
That would allow a "stiff" / "saggy" switch, with the screens on the PT CT and some smallish RC filter OR the screens supplied from B+ the conventional way with a large resistor and a much more squishy reaction to load.
Several years ago I experimented with a circuit I called the sagulator. Instead of a stiff / saggy switch, I had a pot that went from stiff to limp. I used the circuit's output to feed one or more stages in the front end of the amp, the whole preamp, or the screen grid of a pentode somewhere in the chain, often the input or second stage, as high gain (around 1000X) pentodes are usually too microphonic to be the input stage.

The circuit was similar to the diagram in post #2188. I put R3 in the drain of the mosfet and sized it so that the output stayed constant until the load current got into the "damage" region at which point the drain voltage dropped below the gate causing the circuit to drop out of "regulation." As stated, gate protection is required. One end of a pot was tied to the junction of R1 and R2 with the wiper going to the mosfet's gate. The other end of the pot goes to a voltage source that drops when the amp is driven. This can be several places.

In a small amp that will be driven hard, hard enough to blow screen grids, I use a large enough screen grid resistor to put the screen at max rated dissipation when making clipped square waves, and tie the screen side of the resistor to the end of the pot.

In a more mild mannered amp, I used a resistive divider from B+ like R1 and R2, except that R2 is the LDR part of a photoresistive coupler like a Vactrol. The LED in the Vactrol can be driven by any signal that varies like the cathode of the output tube.

Some resistor value tweaking will be needed to keep the sagulator output voltage constant as the pot is turned.
__________________
Tubelab, I blow stuff up so that you don't have to.
  Reply With Quote
Old 22nd November 2021, 01:18 PM   #2194
Tubelab_com is online now Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Enclosed is the schematic for an amp I developed during the active HBAC competition. It just squeaked in under $100 at the time.

The "too many diodes" power supply is both a full wave bridge and a full wave voltage doubler. The power transformer is an ordinary isolation transformer with a single 120 volt secondary. It was chosen for cost reasons. A dual 120 volt secondary transformer with the secondaries in series would make the power supply simpler, much like the one in the Philips schematic.

Either way you get about 160 - 170 volts and 320 - 335 volts from the supply. The 160 volt source feeds the screen grids of the output tubes and powers the heaters. The 320 volt source feeds the output tube plates and the entire preamp.

The output tubes are 45B5 / UL84's which are 45 volt 100 mA versions of the 6CW5 / EL86. The preamp tubes are 26AQ8 / UCC85's.

Ignore the mosfets in the cathodes of the preamp tubes. They were for an early version of the sag circuit which failed. The preamp had far too much gain, so one of the UCC85's was eliminated and the power supply was revised to feed unfiltered but rectified AC to the heaters making live easier on the power transformer. This added yet another diode to the power supply, but diodes are cheap.

This amp made about 20 watts when using a cheap power (mains) toroid for the OPT. I got about 25 watts with a real 3300 ohm OPT, but that didn't fit the challenge budget. I still have it, but prefer the little 4 tube HBAC amp for daily use. It's 4 watts is loud enough.
Attached Files
File Type: pdf HBAC_5T.pdf (29.1 KB, 14 views)
__________________
Tubelab, I blow stuff up so that you don't have to.
  Reply With Quote
Old 22nd November 2021, 11:17 PM   #2195
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Thomas Schwarz View Post
First of all, thank you so much for your very thorough answer!
You're very welcome! In our new Dark Ages of "flat-earthers" and "anti-vaxxers", it's a huge relief to talk about something as real and BS-free as an electronic circuit.
Quote:
Originally Posted by Thomas Schwarz View Post
After reading your comments I looked up the EF80's datasheet - which I clearly should have done before posting my previous remark.
No worries at all. Truthfully, I wanted to draw a schematic for you, so I fired up Kicad and grabbed the first pentode symbol I found in it's library. That just happened to be an EF80!

In the guitar amps I built, the pentodes I used were variously 6AK5, 6AG5, and the pentode section of a 6JW8. At the time, all three of these were available for just $1 USD each. No pricey EF86s for me.
Quote:
Originally Posted by Thomas Schwarz View Post
So a power FET with the voltage fixed from an array of zeners (amplified zener diode) seems to be the way to go -
Well, it's certainly one way to go.

Most guitar players seem to like at least some sag, to "round off the edges" of the sharp transient that starts off every note played on a guitar. If some pentode sag is wanted, a well regulated screen voltage might not be the best option.

As George alluded to, a lot of output tubes in guitar amps die because of too much screen grid dissipation. Sooner or later a guitar amp will be mercilessly overdriven, screen current will soar, and if you don't plan for this, the output tubes will die a fiery death.

Which is why tube guitar amp designers found out, decades ago, that inserting a resistor in series, sized to keep the screens from overheating, should be considered mandatory insurance. Those designers who failed to do this, created unreliable products that failed in use.

Can you imagine a guitar hero playing rock guitar in front of 10,000 people, and suddenly having his guitar amp die in a shower of sparks just as his loud and impassioned guitar solo is reaching its climax?
Quote:
Originally Posted by Thomas Schwarz View Post
A stiff variable screen supply looks like a very attractive way to power-scale the amp ...
...replace R1,R2 with a potentiometer, and you can dial up any screen voltage you want, between 0 volts and B+...

More practically, stick a properly sized pot between R1 and R2 so you can dial up a useful range of screen voltages, neither uselessly low, or uselessly high.

The screen voltage isn't actually regulated in this case (it will track variations in B+), but for a guitar amp, do you really need a mathematically exact regulated screen voltage? Most probably not...

By the way, potentiometers have maximum voltage ratings, and smaller pots invariably have smaller voltage ratings, so do keep that in mind. Also, this is one application where I would choose a plastic-shaft potentiometer over a metal-shaft one, simply for the increased electrical creepage distance between dangerous voltages and your fingertips!

I don't think you mentioned what the amplifier you're thinking about is intended for. Guitar? Something else?

-Gnobuddy
  Reply With Quote
Old 23rd November 2021, 12:07 PM   #2196
Tubelab_com is online now Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
Originally Posted by Gnobuddy View Post
Can you imagine a guitar hero playing rock guitar in front of 10,000 people, and suddenly having his guitar amp die in a shower of sparks just as his loud and impassioned guitar solo is reaching its climax? -Gnobuddy
The Ampeg SVT is a popular bass guitar amp. The new ones are solid state, but the "real" SVT had 6 X 6550's and made a very real 300 watts or more. Unfortunately for Ampeg, there was a first gen SVT which often went down in flames. It used 6 X 6146 RF transmitting tubes and tried to make 300 watts. The 6146 should be treated like a modern sweep tube as far as the screen grid is concerned. If properly fed, the 6146 can crank out 75+ watts per pair even in a bass amp. Ampeg overfed the screens and poof happened.

The most famous flameout was at the infamous Rolling Stones concert at Altamont in 1969 where one of the 6146 based SVT's caught fire. Some unrelated murderous activity also occurred at that show, making the flaming amp or amps a minor event. This wasn't just 10,000 people. Estimates put the crowd at well over 100,000 and parts of the event was televised.

Ampeg blamed it on an incorrect setting of the line voltage switch. Yeah, right, like an amp set on 230 volts is even going to work on 120 volt power. Several other 6146 based SVT's flamed out, and the amps were all switched over to 6550's within the first year.

The 6146 has been regarded as a poor choice for audio ever since, but that is unfounded. This subject pops up here every few years.

6159 for SE Power amp

6146W Power Amplifier. Please help!

8552 finals ?
__________________
Tubelab, I blow stuff up so that you don't have to.
  Reply With Quote
Old 23rd November 2021, 06:34 PM   #2197
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Tubelab_com
...Ampeg SVT...Altamont...
That story was in the back of my mind. It's probably the most notorious case of an imperfectly designed tube amp failing during a live performance, though the tragedy of the four deaths and multiple injuries to people in the crowd completely overshadowed it, as was entirely appropriate.

Unfortunately, music festivals continue to suffer from violence, and violence against women, in particular, tends to be overlooked and under-reported. For instance: Sexual Harassment at Music Festivals Is An Ongoing Problem - FLARE


-Gnobuddy
  Reply With Quote
Old 23rd November 2021, 07:01 PM   #2198
Tubelab_com is online now Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
They got so out of hand in general that the last big festival that I attended was Lollapalooza 96 in West Palm Beach Florida. This was back when Lolla was a touring festival that ran for a whole day on three stages at each stop. The lineup was as diverse as the crowd. Metallica, Soundgarden, Rancid and the Ramones were all in the same show. The punkers played in the afternoon heat and the loud rock / metal came after dark.

There I got to watch a Sovtek MIG 50 go down in flames. Yes, some Russian made guitar amps were sold in the US under the Sovtek brand. They weren't around for long, so I don't know if they all flamed out, or this was a one off crash landing.
__________________
Tubelab, I blow stuff up so that you don't have to.
  Reply With Quote
Old 25th November 2021, 03:51 AM   #2199
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
The Hundred-Buck Amp Challenge
There is a story of an EV 30W playing a show, caught fire, but kept on playing.

That's styrofoam the size of a large bar table.
  Reply to this post

Reply


The Hundred-Buck Amp ChallengeHide this!Advertise here!
Thread Tools

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
Hafler Five Hundred problem hatrack71 Solid State 4 10th June 2009 01:38 PM
most bang-for-the-buck sub-amp skogs Solid State 12 8th September 2007 11:22 AM
Buck & boost transformer Ilianh Parts 1 15th March 2004 07:03 AM
a hundred questions from a noob zoobie Solid State 5 31st May 2003 03:04 PM
Nineteen Hundred Ninety Nine Posts And Counting ;-)... mrfeedback Everything Else 12 21st May 2003 06:59 PM


New To Site? Need Help?

All times are GMT. The time now is 01:59 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 13.64%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Copyright ©1999-2021 diyAudio
Wiki