SE OPT's & Biasing - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

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 27th February 2004, 11:08 AM   #1
diyAudio Member
 
Join Date: Nov 2003
Location: England
Default SE OPT's & Biasing

How much should I read into the wattage ratings given by transformer manucturers?

I want to drive an output stage using a single ended EL34 (6CA7) pentode, a 25 watt tube. I was having a look on Hammond's site earlier and they have something that looks close to what is needed -

1627SE

But they list it's rating as being 25 watts. I will be running the EL34 quite hard, into Class AB. This output stage is for a lead guitar, so tube distortion is actually beneficial. I should really be more specific and say peak-peak clipping, not the roll-off distortion of an OPT going into saturation.

Do you think Hammond mean it will genuinely take 25 watts, or that it will just take 24.999 watts, after which point it melts?

Also, I noticed something odd in a schematic for a guitar amplifier. Rather than bias the power tube, the designer just added a resistance to the cathode of the power tube, set to around the maximum amount of current he wanted to flow through the tube, which was an EL84. Why would this be, and what are the implications of it? If this is fine to do, why not do it for every other stage, after the initial stage, in a common cathode amplifier?

In the preamp, he biases two 12AX7's, so there are four stages. The first is only to lift the guitar's signal to line level. The next stage acts as a volume control, but will begin distorting with hard playing and high volumes. The next is designed to go much futher into distortion, I would guess this would be marked as the 'gain' on the front of the amp.

So he's used 3 stages of a possible 4 and produced a great deal of gain. He then puts the signal thru a cathode follower. Connected to this is a tone stack, a collection of three band pass filters. Would there be any reason for using a cathode follower at this point? It's output resistance for instance? I'm probably making some really dumb mistake, but I don't think I can see any negative feedback being used in this stage.

I was looking at the earlier stages and measured the signal from my guitar while I was playing. I found there was ~100 - 500mv depending on how hard I played, but it averaged at around 300mv. The first stage produces a gain of around 58, which means the singal entering the second stage is around 17.4V right?

If my 12AX7 stage saturates at 0V and cut's off at -2.2V, this means that the signal coming from the second stage could over 100V if the of this stage is roughly the same as the last?
  Reply With Quote
Old 27th February 2004, 11:29 AM   #2
dhaen is offline dhaen  Europe
diyAudio Moderator Emeritus
 
dhaen's Avatar
 
Join Date: Jun 2002
Location: U.K.
Sorry to disappoint, but the EL34 in SE will only give about 10 watts.
As discussed recently, the practical max for class A1 is 25% of anode dissapation. You mention AB, but SE can only run in class A.
It might be possible to push a bit more, but it would mean rather complex drive and power supply circuitry. If you need more, it's simpler to use a pair.

How much headroom transformers give their products varies. I'd happily run a Sowter transformer at, or just beyond it's rated power, and expect good results. I don't know the quality of Hammond. You can tell a fair amount by just comparing weight and dimensions. There can be small variations due to differing materials, but the laws of physics remain the same

SE is particularly sensitive to transformer ratings. Too small a transformer will spoil the LF response, causing level deficiency and distortion.
  Reply With Quote
Old 27th February 2004, 11:49 AM   #3
EC8010 is offline EC8010  United Kingdom
diyAudio Moderator Emeritus
 
EC8010's Avatar
 
Join Date: Jan 2003
Location: Near London. UK
Output transformer specifications are a very variable quantity, and frequently need to be taken with a pinch of salt. Maximum power capability is inversely proportional to frequency and distortion, so any transformer will distort if you drive it hard enough or low enough. Worse, inadequate primary inductance causes the valve to distort as it has to provide current in addition to intentional load current. The way to avoid this is to specify the primary inductance to have a reactance equal to the load resistance at the lowest frequency of interest. Thus, the Williamson specified 100H for a 10ka-a load, making it good down to 16Hz. Achieving this sort of inductance and a sensible flux density in the core usually requires a large core, making for a big, heavy transformer as dhaen has already pointed out.

An electric guitar is a very boring instrument played through a Hi-Fi amplifier. It's the guitar amplifier's distortion that gives the entire instrument its character. Normal rules of Hi-Fi design need not apply.
__________________
The loudspeaker: The only commercial Hi-Fi item where a disproportionate part of the budget isn't spent on the box. And the one where it would make a difference...
  Reply With Quote
Old 27th February 2004, 01:07 PM   #4
diyAudio Member
 
Join Date: Nov 2003
Location: England
Quote:
Sorry to disappoint, but the EL34 in SE will only give about 10 watts.
It's not really a disappointment, rather a slight benefit. I specifically chose think about SE designs as 50 watt PP design's are far too loud to run them all out, especially around the home.

My amp is a 100 watt half stack. I bought this not quite knowing the meaning of what loud was. I have had complaints from neighbours without it ever reaching 5 on it's volume. Above 5, it's unrealistic. The window panes shake, my ears feel like they're about to bleed and the police are ready to visit.

At the moment, my family complain if it ever goes above 2. So power tube distortion is never going to happen. In fact, the amp barely even seems to be alive most of the time. Which is a real annoyance for me.

Quote:
You mention AB, but SE can only run in class A. As discussed recently, the practical max for class A1 is 25% of anode dissapation.
I reversed your points to help me understand them. Is this due to the quiescent current continuously heating the anode even while no signal is being amplified? 10W is ~4W over 25% isn't it? I guess this is because it's not true class A1?

Quote:
If you need more, it's simpler to use a pair.
I may end up going that way. I was reading about how SE amplifiers have problems with OPT's causing distortion, but, similarly, a while ago I was reading about how it can be very easy for a PP's phase splitter to distort the signal. So I've started to see it as a trade off. With a PP design, I would need to start using phase splitters, while an SE leaves me with the OPT to deal with.

I also noticed a lot of high end systems seem to believe the phase splitter causes more distortion of the signal than the effects of the OPT in an SE amp.

Quote:
You can tell a fair amount by just comparing weight and dimensions. There can be small variations due to differing materials, but the laws of physics remain the same
I read that a couple of manufacturers use ?pinstripped? laminations with mu-metal every other iron lamination to try and remove none linear coupling. I don't know how true this is or how widely it's used. Mu-metal has a very high permeativity, so it should begin coupling earlier and then saturate soon after right? Which would help couple the lower frequencies?

Quote:
An electric guitar is a very boring instrument played through a Hi-Fi amplifier. It's the guitar amplifier's distortion that gives the entire instrument its character. Normal rules of Hi-Fi design need not apply.
Unless you like clean guitar's I guess! But I see what you mean. At the same time, there are certain forms of distortion that I don't want to hear. A lot of amplifiers have a very rounded off sound to them. While they are clipping the signal, they are also smoothing off some parts dependant on their frequency. So every guitar starts to sound the same as the rest, and looses it's harmonic signature.

The simplest example I can think of is rectifiers. Mesa Boogie often use tube based rectifiers in their lead amplifiers. They sound okay to me, but they seem to mask the real dynamics of the player in voltage rise times. Where as Marshall use almost only solid state rectifiers, which sounds much more upfront. If you compare Metallica to Van Halen, that's a good example of what I mean!

I don't mind tube's clipping the signal, but such frequency dependant rounding off I dislike.
  Reply With Quote
Old 27th February 2004, 01:20 PM   #5
AndyN is offline AndyN  United States
diyAudio Member
 
Join Date: Oct 2003
Location: Marin County, CA
Default Hammond 1627s

I've used Hammond 1627s in an SE 6L6, and boy are they big. I think they are 5 Kg each. In this case I'd expect that if they are advertised as handling 25 W, they will - advertising optimism aside.

I am not a guitarist, but I was under the impression that trafos for guitar use were often intentionally undersized - the resulting distortion being a major component of amp "tone".

If you are looking for that big fuzzy blues sound, I'd think an over-qualifed transformer is not what you want. Playing electrified acoustic might be alright.

And simply from a logistical standpoint - Do you want to lug an amp w/ a 5 Kilo OPT around?


On preview,

>I don't mind tube's clipping the signal, but such frequency >dependant rounding off I dislike.

The interwinding capacitance of the trafo is going to be an issue at HF. A bigger winding will have more capacitance, and more HF rounding.

Have you looked at the SE guitar amps that Wavelength Audio builds? http://www.guitar-engines.com/index.html Gordon Rankin may have some ideas he'd be willing to share.

Andy N
__________________
The oboe, it is clearly understood, is an ill wind that no one blows good.
  Reply With Quote
Old 27th February 2004, 01:55 PM   #6
diyAudio Member
 
Join Date: Nov 2003
Location: England
Quote:
I am not a guitarist, but I was under the impression that trafos for guitar use were often intentionally undersized - the resulting distortion being a major component of amp "tone".
For some players it is, but I don't particularly like it. It's described as squish, or sag. Like rectifiers, to me it sound's okay, but I'd rather loose the slight rounding if it allows me to hear how my guitar it's self sounds - to let some more of the harmonics come through.

The players who use tube amps always accuse the ones using the newest digital modelling amps of playing with a computer. Not quite so directly, but it's behind almost everything they say towards each other.

But in a similar way, heavily saturated tube amplifiers seem to be as equally disconnected to me. It feels more like you're playing a note, and the amplifier is making a corresponding sound, not amplifying what you've played.

The newest bands on TV are really bad for using this style of playing. They not only use super squishy amplifiers, but then use horrible amounts of clipping to help cover over any mistakes and make it sound like any modern, society hating band should sound.

I don't like fully distorted, and I don't particularly like clean sounds. So I want to kind of going for the middle on this one.

Quote:
And simply from a logistical standpoint - Do you want to lug an amp w/ a 5 Kilo OPT around?
Actually, it's not so bad. I used to play with a few friends. The drummer always assumed that unless there was going to be a drum set already there, whenever we played we'd be at his house.

It's not the amplifier it's self that's a pain in the *** to move, it's the 3ft tall, 50kg speaker cabinet that goes with it. Along with the associated miles of cabling and adaptors.

I got sick and tired of the 'I have a drum set so I can't move it' excuses and gave up dragging it round there. It's more of a physical volume problem, since most 4 x 12 cabinets won't fit in a car very easily. He also had a very small set of stairs up to where we practiced, the cabinet barely made the corner in them.

It's exactly like that say. These guys I played with are two of the best friends I know. Yet the minute we started playing, all the issues come out and people start getting very angry, very very quickly. It took me about two or three times playing with these guys to get so annoyed that I didn't want to continue.

Quote:
The interwinding capacitance of the trafo is going to be an issue at HF. A bigger winding will have more capacitance, and more HF rounding.

Have you looked at the SE guitar amps that Wavelength Audio builds? http://www.guitar-engines.com/index.html Gordon Rankin may have some ideas he'd be willing to share.
I'll check it out. I know that the highest fundamental frequencies of a lead guitar are around 1 - 1.5kHz depending on which tuning you decide to use. So I would expect the third harmonics to be around 12kHz, which is where I saw such SE OPT's from Hammond beginning to have problems with overshoot and distortion.

The good thing is that the highest notes of a guitar are very rarely the majority of a tune. Almost all verse & chorus riffs are around the low E6 string, which has an open fundamental frequency of ~100Hz, again, tuning dependant. SE still has a chance!

Awww... I feel sorry for it now.
  Reply With Quote
Old 27th February 2004, 03:06 PM   #7
AndyN is offline AndyN  United States
diyAudio Member
 
Join Date: Oct 2003
Location: Marin County, CA
Quote:
Originally posted by eeka chu
The players who use tube amps always accuse the ones using the newest digital modelling amps of playing with a computer.
My buddy plays through a POD, into a Mesa Boogie MKIIc. He loves it; I'd rather build amps.

Quote:
I know that the highest fundamental frequencies of a lead guitar are around 1 - 1.5kHz depending on which tuning you decide to use. So I would expect the third harmonics to be around 12kHz, which is where I saw such SE OPT's from Hammond beginning to have problems with overshoot and distortion.

The good thing is that the highest notes of a guitar are very rarely the majority of a tune. Almost all verse & chorus riffs are around the low E6 string, which has an open fundamental frequency of ~100Hz, again, tuning dependant. SE still has a chance!
Looks like you're registered in the UK, so I'm not sure if this is a practical idea, but here goes: You might look at Electra-print transformers. Jack Elliano, the owner, loves to talk at length about the various aspects of transformer design. He does custom work, his prices are reasonable and his quality is great. The money invested in an hour-long phone talk with Jack wouldn't be wasted. Hs is in Las Vegas, which is GMT-8 time I think. Plus the dollar is weak now; never been a better time to buy. I'm just a happy customer, no affiliation.

Do you feel comfortable with the distinction between 'traditional' Direct Feed and the 'old-is-new-again' Parallel Feed? (aka parafeed, or shunt feed) I get the feeling that you'd like the parafeed topology. Gordon Rankin has a good write-up here: http://www.guitar-engines.com/ampspf.html

The key point is that parafeed allows you to use the exotic high-permeability metals in the OPT, since they are relieved of the job of passing DC current. DC saturates the exotic stuff, rendering it useless. Like you've mentioned, the hi-perm stuff is desireable because it works better on the low end. Plus you can wind on a smaller core, allowing for lower interwinding capacitance = better high end. Win/win, if you can afford it & are happy with under 10W or so.

As a point of reference, I listen to a homebrew Single-Ended parafeed 2A3 amp with a Magnequest cobalt OPT (cobalt is ultra hi-perm). 2 watts of wideband low-distortion bliss. You've got me thinking... Someday I've gotta invite my pal to bring one of his pickup equipped acoustic guitars over to play through it.

Cheers, Andy
__________________
The oboe, it is clearly understood, is an ill wind that no one blows good.
  Reply With Quote
Old 27th February 2004, 03:17 PM   #8
AndyN is offline AndyN  United States
diyAudio Member
 
Join Date: Oct 2003
Location: Marin County, CA
Default Bias

Did you get an answer to your question about Bias? The resistor under the cathode?

It's called self-bias. The resistor elevates the voltage of cathode by the desired amount, and the grid is tied to ground via a few-Kohm resistor. The grid is at ground potential, which means that it's negative relative to the cathode, which (voila!) is a biasing difference. It's a nice technique because it obviates the need to build a seperate bias supply, and it automatically adjusts the bias as the tube ages.

The cathode resistor is often bypassed by a largish capacitor, that shunts AC frequencies to ground. It's also a bit of negative feedback (I'm told) which raises gain a smidge. Some people advocate removing that cap; it works both ways, and is a sonic taste issue. I leave 'em in.
__________________
The oboe, it is clearly understood, is an ill wind that no one blows good.
  Reply With Quote
Old 27th February 2004, 04:01 PM   #9
diyAudio Member
 
Join Date: Feb 2004
Location: Zinzinnati
Default Guitar Amps info

Gang & John,

Just a couple of things...

First SET pentodes do work great with guitar amps. I will be releasing a Rob Fetters signature 25W 6550/KT88 amp head/combo soon.

The reason for the smaller output is because the low end is equated to low E or about 80Hz. Going to 20Hz requires much more inductance and therefore more iron. Though less iron is not the way to go as many of these output transformers available for SET guitar amps are too small and require too much feedback.

Ok a word about gain... You will need a ton of gain in a tube guitar amp. Think of it this way... Max input is usually speced at 10mv. The Tone Stack robs between 20-30dB (ouch) and the EL34 for 10W output requires at least 10V's input (assuming 5K load). Ok so you will need at least 90dB of gain or a gain of 30K going to into the EL34.

On gain... 12AX7A into a Tone Stack lucky if you get a gain of 45. As far as a driver stage 12AX 65 and remember the needed feedback usually -3dB. So you really need like a 12AU7 stage with a gain of 10-12 to get enough drive into the EL34 to get it too 10W's.

SET's do sound really great for guitar amps in many ways and sound much larger than their PP counterparts by a factor of 3-4 times.

But that's only my opinion others always welcome!

Thanks
Gordon
__________________
J. Gordon Rankin
Wavelength Audio, ltd.
  Reply With Quote
Old 27th February 2004, 04:48 PM   #10
diyAudio Member
 
Join Date: Nov 2003
Location: England
Quote:
He loves it; I'd rather build amps.
I hate the mathematics and complex physics involved in so much of building electronics well.

At the moment, my main drive to build my own amp is that I'm not happy with what I have at the moment. It's far too loud and now quite detailed enough for me.

I like listening to music, but it's nothing compared to standing in front of the the speakers and knowing it's you making the music!

Quote:
Do you feel comfortable with the distinction between 'traditional' Direct Feed and the 'old-is-new-again' Parallel Feed? (aka parafeed, or shunt feed) I get the feeling that you'd like the parafeed topology. Gordon Rankin has a good write-up here: http://www.guitar-engines.com/ampspf.html
I started to think about this just before I checked back here. I've never really looked at it, but my initial idea was -

"Hey! Wouldn't it be excellent if I could just drop the DC away from the transformer through some kind of broadband high pass filter!"

When I first saw the idea on Gordan's site, it looked a bit weird, but I understand it fine now.

I'm not so sure of the real world component costs and realities though, as it is mentioned that the idea only becomes realisitically usable under 8W and above 16W.

Quote:
Did you get an answer to your question about Bias? The resistor under the cathode?
As soon as you began explaining it I realised what had gone wrong, and it was me not noticing the ground reference resistor connected from the grid to to ground!

Rather than being squashed up close to the tube, it was shunted over, futher away, so from the corner of my eye I must have thought it was a volume control.

Not enough coffee...

Interesting, the particular design in question uses an EL84 in a SE design and suggests a dissipation of 11.7W? This design definitly doesn't use any special arrangements for it's output.
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Biasing fauxpas Tubes / Valves 4 3rd January 2008 08:06 AM
low-biasing or high biasing in Class-AB amps Workhorse Solid State 124 18th October 2005 04:45 AM
Biasing Wyatt Turk Tubes / Valves 10 1st October 2004 02:11 PM
SOZ Biasing vdi_nenna Pass Labs 21 2nd June 2001 03:38 AM


New To Site? Need Help?

All times are GMT. The time now is 04:11 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2