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.

OTL amps for guitar duty
OTL amps for guitar duty
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
Closed Thread
 
Thread Tools Search this Thread
Old 21st February 2019, 10:53 AM   #181
Printer2 is offline Printer2  Canada
diyAudio Member
 
Join Date: Apr 2010
Would a 7695 cut it for 32 ohm? I got four. Other than that a bunch of 12L6's or 6CW5's are about the only higher current tubes I got. Not really interested in getting rid of the OT as if it was weight that was an addition the two pounds is not a big deal once you get the power transformer in the switching realm. Had a thought, but wouldn't you need to compare the same tubes going through a transformer to really compare the two?

Last edited by Printer2; 21st February 2019 at 11:10 AM.
 
Old 21st February 2019, 01:44 PM   #182
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
The 7695 and related 6 volt flavor the 7754 were designed for audio use, and primarilly for the "600 ohm OTL" circuit on the last page of the data sheet. Being an "audio tube" there is no hard spec for maximum cathode current. I have one of each, so I haven't tried them in anything.

Connecting a speaker directly to a vacuum tube is like trying to drive your car 30 MPH in 5th gear. The engine (tube) needs to be running at enough RPM (voltage) to flow air properly and make enough TORQUE (move enough current) to do it's job. The transmission and differential (OPT) multiplies torque (current) at the expense of RPM (voltage).

We can find a big enough engine (tube) to make the required torque at a low RPM, but it will be operating outside of its design parameters and will dissipate a lot of its energy input as waste heat.

So, lets find our big engine that will provide the lowest losses when operated at a low voltage and high current. Here we want 10 watts into 32 ohms. That's roughly 18 volts RMS or 50 volts peak to peak across the load. Peak current into a resistive load should be about 800 mA, but slamming a cone reversing transient into a speaker (say popping the clutch) will need more current.

We only need 50 volts of signal swing, but most tubes need 50 to 100 volts of headroom to get this. Few tubes can pull their plates below 50 volts at the typical current they see, but now we want a low saturation voltage at nearly an amp of current. I'm not sure what the 7965 will do when asked for this. Some tubes that I have tested will arc from plate to cathode resulting in tube death. Others will simply glow red and not pass the current (not enough cathode emission). Here multiple tubes in parallel are required.

As the OTL people have discovered over the years the big regulator triodes are the best tube for this job. They range from tiny (6S4 and 12B4) small (7233 and Russian 6C19) mid sized (6AS7, 6080, 5998 and 7236) big (6336 and 6528) and HUGE (7241 and 7242). Some of these are extinct and expensive.

When I was searching for the best tube in a cathode follower amp which used a modulated B+ supply (class H) I needed big current at low saturation voltages. Some big TV sweep tubes are better suited for the job than the regulator triodes, but require the added complication of pentode operation. I used a big regulator triode for that amp design for simplicity.

I have over 100 25DN6 tubes that cost me $0.50 each, so I may experiment with them, but look at these curves for a BIGGER sweep tube. Yes, that's over 1 amp of plate current at 50 volts on the plate. Short tests (about 500 mS) to nearly 1 amp were done at 650 volts. Rated CONTINUOUS cathode current is 400 mA and the peak rating is 1.4 amps. That can slap some speakers around.

I am including the magazine article I published about 10 years ago that was the result of all this cathode follower stuff. It was removed from Circuit Cellar's web site long ago.
Attached Files
File Type: pdf LW6_tests.pdf (86.1 KB, 3 views)
File Type: pdf 2910014_Anderson.pdf (289.1 KB, 6 views)
__________________
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......
 
Old 21st February 2019, 03:45 PM   #183
45 is offline 45  Italy
diyAudio Member
 
45's Avatar
 
Join Date: Dec 2008
Location: UK
I think that a sweep pentode, in pentode operation, is likely the best choice for guitar amp OTL. The high Zout is no problem. It actually helps with big cones.....
The two extremes are Fender and Vox. The former uses pentodes with fbk to get a clean sound and the latter uses pentodes without fbk to get a sound that is never really clean.
I use pentode operation without feedback in all my amps and I am able to get super clean output just staying in class A or very near. Plate efficiency is a bit lower than usual but still better than triodes....up around 40-50% with an output transformer.

Last edited by 45; 21st February 2019 at 03:47 PM.
 
Old 21st February 2019, 04:15 PM   #184
Printer2 is offline Printer2  Canada
diyAudio Member
 
Join Date: Apr 2010
Yeah I knew the tubes I got are kind of limited but I am not really looking for 20W, 1W might be fine just to mess arround with paralling up a bunch. Or maybe not. I'll break down and buy a TV tube or two yet, it is just the shipping that bugs me.
 
Old 21st February 2019, 07:59 PM   #185
atmasphere is offline atmasphere  United States
diyAudio Member
 
atmasphere's Avatar
 
Join Date: May 2007
Quote:
Originally Posted by 45 View Post
Hey George, you take it as a challenge and that 's fine.

Someone else is certain but cannot make an example. That's not fine.
Calling you out on that one. My surmise is you didn't look at the link I posted earlier.

I got a pair of power transformers for the output at a local surplus outlet (Acme Electronics, defunct since losing their lease) for $6.00. I can't tell what brand but they are marked 'A10310B' and have 117V in and 92.5Volts out. Later I added a control transformer to buck the AC another 12 volts for a little more power.

The 6080s were obtained at a ham swap meet (Midwinter Madness although it was in March) for $2.00 each. 6080s usually don't hold up as well as 6AS7Gs but for $2.00 each I wasn't complaining. A 6SN7 was sufficient to drive them. Sockets came from a coffee can full of used sockets in my basement. I might have $5.00 into that can; had to clean up the sockets prior to use. I found a filament transformer (Signal Transformer Co.) that did 10amps for $10.00, this time from Ax-Man Surplus. A Bud chassis served; $15.00. I used one of our driver power supply transformers for the 6SN7 B+ and B- (I ran it as a cathode follower). That transformer was $12.00 when new (its been sitting on the shelf here for about 20 years). Two 12AX7s suited as the voltage amplifier and preamp. I suppose I have another $20.00 in junkbox parts to put it together with a switch and fuses, Alps controls and Switchcraft 1/4" phone jack. Filter caps gobbled a bit of the budget but no worries coming in under $100 total let alone 100 Euros.

The circuit runs zero feedback. I'd need another gain stage to add feedback. That's why a higher impedance load works best. Otherwise it tends to be too bassy. A box with several drivers in a line source array works pretty nice as the sound pressure does not drop off so quickly as you move away from it.

In my regular amps (production prototypes with more finished chassis work) I've made two versions (each using four power tubes). One uses a direct coupled cathode follower as the driver similar to the prototype above. The other uses an Edcor transformer to convert from single-ended to balanced output for the grids of the power tubes.

Since I'm not running a whole lot of bias current, I don't have either a bias or DC Offset control since the DC Offset current is minimal if the tubes are OK. Its not class A, but it has no crossover artifact since there is no magnetic field to collapse.
 
Old 21st February 2019, 09:24 PM   #186
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
1W might be fine
I think that you can get there from here.

The 7695 physically looks like a 6W6 / 12L6 type with added fins on the hot sides of the plate for a bit more dissipation. Typical operation has it idling at 100 mA in SE class A. The heater is also 7.5 watts, so maybe it can do the same thing as a 12L6 which is rated for 60 mA average / 180 mA peak in TV sweep use. Assuming that any of these can do 100 mA idle, 200 mA peak, a pair in parallel can put enough current into 32 ohms for a bit over 1 watt.

1 watt into 32 ohms needs 5.7 volts RMS at 175 mA. Peak currents should be in the 250 mA range and peak to peak voltage swing should be around 16 volts, but that's driving a resistor. A 8 ohm guitar speaker near resonance can be 20 ohms or more, and resonance is often within the guitar's frequency range for smaller speakers. I imagine any B+ voltage that gets the tubes to work will provide plenty of available voltage swing.

The (conservative) 10 watt dissipation rating comes into play as B+ goes over 100 volts on the 6W6 / 12L6 if you want to stay in class A. The 7695 can eat 16 watts.

There are several different circuit topologies that can be used for OTL. The totem pole like SemperFi's design, the Circlotron like Atmasphere uses, the Futterman, and the SRPP like circuit shown in the 7695 data sheet.

It has been postulated that an SE OTL would be the worst of both worlds, but it just might work for a guitar amp, and it offers some other circuit choices like the SRPP with some minor tweaks.

Why would I go there.....because I, and others, have been there....a long time ago. Some of the experiments are on a forgotten page in my web site:

Active Loaded SE Output Stages | Tubelab

and the link leads back to some experiments here:

Active load output stage

and to an amp by Pete Millett that looks like an EL34 version of the SRPP thing on the 7695 data sheet.

EL34 Active-load (SRPP) amp

Granted all of these experiments were done with OPT's or big resistors, but the concept can be applied to OTL. All of the circuits are similar with minor tweaks. I would probably use a zener to control the screen grid voltage in pentode mode.....but maybe not....I'll have to try it.
__________________
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......
 
Old 22nd February 2019, 08:55 AM   #187
45 is offline 45  Italy
diyAudio Member
 
45's Avatar
 
Join Date: Dec 2008
Location: UK
Quote:
Originally Posted by atmasphere View Post
Calling you out on that one. My surmise is you didn't look at the link I posted earlier.
I can't see anything but things related to music.

Quote:
Originally Posted by atmasphere View Post
I got a pair of power transformers for the output at a local surplus outlet (Acme Electronics, defunct since losing their lease) for $6.00. I can't tell what brand but they are marked 'A10310B' and have 117V in and 92.5Volts out. Later I added a control transformer to buck the AC another 12 volts for a little more power. The 6080s were obtained at a ham swap meet (Midwinter Madness although it was in March) for $2.00 each. 6080s usually don't hold up as well as 6AS7Gs but for $2.00 each I wasn't complaining. A 6SN7 was sufficient to drive them. Sockets came from a coffee can full of used sockets in my basement. I might have $5.00 into that can; had to clean up the sockets prior to use. I found a filament transformer (Signal Transformer Co.) that did 10amps for $10.00, this time from Ax-Man Surplus. A Bud chassis served; $15.00. I used one of our driver power supply transformers for the 6SN7 B+ and B- (I ran it as a cathode follower). That transformer was $12.00 when new (its been sitting on the shelf here for about 20 years). Two 12AX7s suited as the voltage amplifier and preamp. I suppose I have another $20.00 in junkbox parts to put it together with a switch and fuses, Alps controls and Switchcraft 1/4" phone jack. Filter caps gobbled a bit of the budget but no worries coming in under $100 total let alone 100 Euros.
If the cost of a 50-100 VA transformer is $6 then my amp is quite cheaper than $100. I have made the PSU myself and thought it would cost $30-35. It's highly interleaved for ultra-low capacitance...not the usual cheap stuff....

To make a 16K transformer you can use the Edcor XPP15-8K with 4 ohm seconday and use it with a SINGLE standard (and good) 8R driver...!!. It costs $22.67. This is rated 15W I suppose at the US line frequency of 60Hz which means about 27W at 80Hz if 8K and over 13W if 16K. So it's good for both the 10W Class A PCL86 amp and the 20W EL84 amp! $30 max for chassis and wooden box and that's pretty much where big money is.....EL84's are cheaper than one might think just selecting a variant which is actually better because of the different pin out which allows higher ratings for plate voltage and g2 voltage to make a very reliable 20W amp. IT also allows to simplify the bias supply too....!! PCL86's like Polamp (old Philips Polish factory) I paid them $0.5 each.


Quote:
Originally Posted by atmasphere View Post
The circuit runs zero feedback. I'd need another gain stage to add feedback. That's why a higher impedance load works best. Otherwise it tends to be too bassy. A box with several drivers in a line source array works pretty nice as the sound pressure does not drop off so quickly as you move away from it.
Already a compromise.....How many drivers??? A 6" cone cannot compete with 10" or 12" cones. With an array of 6" cones it has to be seen, interference will happen for sure, so the sound will be quite dependent on the listening position. This might work under certain conditions (compromise) but it might also not work at all for more than someone.....

Quote:
Originally Posted by atmasphere View Post
In my regular amps (production prototypes with more finished chassis work) I've made two versions (each using four power tubes). One uses a direct coupled cathode follower as the driver similar to the prototype above. The other uses an Edcor transformer to convert from single-ended to balanced output for the grids of the power tubes.
What's the point of all your OTL theories if you use a CHEAP signal transformer anyway???

Quote:
Originally Posted by atmasphere View Post
Since I'm not running a whole lot of bias current, I don't have either a bias or DC Offset control since the DC Offset current is minimal if the tubes are OK. Its not class A, but it has no crossover artifact since there is no magnetic field to collapse.
Still that is another weakness and not really recommendable given what people do with a guitar amp. What happens if the output tube fails?? Because guitar amps are not HiFi amps if they are really used sooner or later output tubes will fail or have to be changed before it's too late.... I have seen enough to be certain that people do not care about checking the status of the output tubes routinely.....
There is no collapse in a properly made output stage with OPT. Also because for best performance the output stage shouldn't clip heavily. Distortion can be better made upstream....
I have actually found a way to wind a relatively inexpensive transformer for unity coupling and thus class B operation just like the McIntosh! For those who can't the twin coupled amp as proposed by Crowhurst is a good solution.

Last edited by 45; 22nd February 2019 at 09:22 AM.
 
Old 22nd February 2019, 11:09 AM   #188
45 is offline 45  Italy
diyAudio Member
 
45's Avatar
 
Join Date: Dec 2008
Location: UK
Oh I see the picture on Fbk. Sorry I do not have an account and never will.

Anyway if that sort of ship is meant to be a guitar amp you will sell none or near none. The chassis doesn't look light at all. Where is the weight saving? A typical 50W amp with EL34's is a lot more compact and can be possibly lighter than that.
A guitar amp has to be practical. This is a crucial requirement just like the sound! We are not talking about stage amps with multiple arrays as large as a house.
50W on which load? 32R? 64R? more? How do you put the drivers? Are you aware of interference and limited freedom in placing them in order to get best result? If you use 2x16R drivers they cannot be too small and if use 4 putting them into a square or similar is not a good idea. I know this has be done and can still be found but it's not a great idea....
 
Old 22nd February 2019, 04:53 PM   #189
atmasphere is offline atmasphere  United States
diyAudio Member
 
atmasphere's Avatar
 
Join Date: May 2007
Quote:
Originally Posted by 45 View Post
I can't see anything but things related to music.
The first link was to my regular website. The second link was to my band.

Atma-Sphere

Quote:
Already a compromise.....How many drivers??? A 6" cone cannot compete with 10" or 12" cones. With an array of 6" cones it has to be seen, interference will happen for sure, so the sound will be quite dependent on the listening position. This might work under certain conditions (compromise) but it might also not work at all for more than someone.....
If you place them close to each other, and have at least 4, you form a line source array- no cancellation. I used Jensen 8" units and they work quite well.

Quote:
What's the point of all your OTL theories if you use a CHEAP signal transformer anyway???
To get more of the kinds of distortion made by regular guitar amps without the weight and cost. It works pretty well in that regard- plenty of bandwidth and no worries driving the power tubes into saturation. I run a single-ended front end clear up to the driver tube to get a bit of 2nd-order harmonic richness. Sunn used to make solid state amps that were a similar topology and have a reputation as some of the richer solid state amps made. Their driver transformer wasn't a miracle device either


Quote:
Still that is another weakness and not really recommendable given what people do with a guitar amp. What happens if the output tube fails?? Because guitar amps are not HiFi amps if they are really used sooner or later output tubes will fail or have to be changed before it's too late.... I have seen enough to be certain that people do not care about checking the status of the output tubes routinely.....
There is no collapse in a properly made output stage with OPT. Also because for best performance the output stage shouldn't clip heavily. Distortion can be better made upstream....
If an output tube fails you replace it. Pretty much like other amps in that regard


The amp in the photo on our FB page is a prototype. The chassis was designed to be mounted on its side in a combo. But much later on I built it up as a guitar head- we were going to put it upside down in an open back using one or a pair of 10" drivers originally meant for an SVT cabinet.

Last edited by atmasphere; 22nd February 2019 at 04:59 PM.
 
Old 22nd February 2019, 07:05 PM   #190
45 is offline 45  Italy
diyAudio Member
 
45's Avatar
 
Join Date: Dec 2008
Location: UK
Quote:
Originally Posted by atmasphere View Post
The first link was to my regular website. The second link was to my band.

Atma-Sphere
I knew your website but cannot see any guitar amp there.

Quote:
Originally Posted by atmasphere View Post
If you place them close to each other, and have at least 4, you form a line source array- no cancellation. I used Jensen 8" units and they work quite well.
Rather simplistic assumption. From my past experience you will have interference in high midrange and treble range for sure.
The only full range driver line array that works well is that one using 4" cones MAX size. 5" is already too big! And it has to be rather high to approximate a cylindrical line source. In your 4x8" cones emission lobes will develop.
Moreover its an array so NO way of getting a reasonably sized and/or portable amplifier. Totally unpractical. Only usable if never moves otherwise it's just for museum display.
I am curious to see where do you put the amplifier then. I doubt you can make a reasonable combo amp.

Quote:
Originally Posted by atmasphere View Post
To get more of the kinds of distortion made by regular guitar amps without the weight and cost. It works pretty well in that regard- plenty of bandwidth and no worries driving the power tubes into saturation. I run a single-ended front end clear up to the driver tube to get a bit of 2nd-order harmonic richness. Sunn used to make solid state amps that were a similar topology and have a reputation as some of the richer solid state amps made. Their driver transformer wasn't a miracle device either
Without the weight and cost? You cannot make such general statements sorry. You just wrote that for the 50W amp amp you made an array of 4x8" cones. + head. That is heavier than a 50W combo amp with its OTP and a single driver. No question.
You missed my point. You have been raving about the transparent sound of OTL amps and then you use a cheap interstage transformer that is typically worse than an output transformer of the same price range! You confuse transformer saturation and power stage clipping. One doe not imply the other.

Quote:
Originally Posted by atmasphere View Post
If an output tube fails you replace it. Pretty much like other amps in that regard
You might need to replace the speakers too....besides tubes in OTL amps, as used in guitar world, have surely shorter life than typical pentodes. So the long term cost quite higher.

Quote:
Originally Posted by atmasphere View Post
The amp in the photo on our FB page is a prototype. The chassis was designed to be mounted on its side in a combo. But much later on I built it up as a guitar head- we were going to put it upside down in an open back using one or a pair of 10" drivers originally meant for an SVT cabinet.
2x10" its not small. For a 50W amp might be fine but a 10-20W amp? Surely not.
If OTL guitar amps had something more to offer or were more convenient be certain that you would not be alone....doing it for fun is one thing but it ends there.
 

Closed Thread


OTL amps for guitar dutyHide 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
OTL guitar amp... M Gregg Instruments and Amps 6 24th May 2017 09:04 AM
IS there any OTL pre amps out there? djn Tubes / Valves 36 15th December 2010 08:37 PM
Anyone out there try a SEPP OTL guitar amp?? aletheian Instruments and Amps 10 14th October 2005 03:05 AM
heavy duty valve amps.... for PA use....? jc808 Tubes / Valves 29 1st January 2005 11:40 PM
[INDIA]Building heavy duty Chasis for Power Amps. ajju Parts 11 20th November 2004 03:55 AM


New To Site? Need Help?

All times are GMT. The time now is 08:52 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki