[Steve Eddy]

What I'd like to test for is whether the OTL amp has anything resembling approximate power source or current source characteristics.

Well that would be determined by the amplifier's output impedance, which you have already said is 7 ohms.

I'd measure a single driver driven by the OTL amp at a fixed input voltage, and then measure two such drivers wired in series.   If the amp has a voltage-source-plus-series-resistance characteristic, the broadband SPL should be unchanged.  If for some reason the amp kinda sorta somewhat approximates a power source or even a current souce, that should be obvious from a significant increase in SPL with the two drivers wired in series.  I could test the parallel connection also, anticipating less than the voltage-source-predicted 6dB increase in SPL.

I could run the same test with a solid state amp as a control.

Assuming I keep all input levels the same and take care with microphone setup and use decent measuring equipment, does this sound like a reasonable test?

With all due respect, I'm not sure you quite understand what's going on here and in spite of what I've said in previous posts you seem intent on making all of this much more complicated than it is and seem to be looking to uncover some mystery that's not there so I'm just going to bow out of this discussion.

se

[Duke]

Steve,

Sorry I chased you off.  I hope I wasn't abrasive, but I can understand an engineer becoming frustrated by a non-engineer's failure to understand.

Let me ask you one question before you go, if you don't mind:

In this context, is "very high output impedance" (like seven ohms ballpark - I'm not sure that's the exact number) another way of saying "kinda-sorta-somewhat current-source-ish"?

It sounded to me like you were saying the amp I described has the one set of characteristics without the other, but maybe they are just different ways of saying the same thing?

Okay, I guess that was two questions, or maybe just the same one twice.

Thanks,

Duke

[Steve Eddy]

Sorry I chased you off.  I hope I wasn't abrasive, but I can understand an engineer becoming frustrated by a non-engineer's failure to understand.

No, you weren't abrasive. Just that you kept going on about the amp acting like a power source and power being the more meaningful measure, etc.

Let me ask you one question before you go, if you don't mind:

Ok.

In this context, is "very high output impedance" (like seven ohms ballpark - I'm not sure that's the exact number) another way of saying "kinda-sorta-somewhat current-source-ish"?

Well, at 7 ohms, I'd say instead of being "kinda-sorta-somewhat current-source-ish," it's really just a rather poor voltage source.

It sounded to me like you were saying the amp I described has the one set of characteristics without the other, but maybe they are just different ways of saying the same thing?

My advice would be to forget about trying to define the amplifier's output as voltage source-ish, current source-ish, or power source-ish and getting yourself distracted by a tangle of semantics and just focus on the amplifier's output impedance and its implications for driving loudspeakers.

You don't need much more than Ohm's Law and know how to calculate voltage dividers. For the speakers, you can use the amplifier's output impedance to recalculate the speaker's Qts.

First calculate the new Qes by way of (Re + Rs) / Re where Re is the speaker's voice coil DC resistance and Rs is the amplifier's output impedance.

Then with the new Qes you can recalculate Qts by way of (Qms x Qes) / (Qms + Qes).

Use some speaker CAD software to get the low frequency response of the speaker with its modified Qts compared to the same speaker using the published Qts.

se

[doug s.]

i also prefer tubes on top, & solid state down low.  but, for me, whether or not you can bi-amp your speakers is a moot point.  i am happy to run tubes above ~80hz, & use solid state on subwoofers, actively crossed over to my monitors, which see only tubes, whether or not they can be bi-amped.  which tube amp to use is a matter of speaker matching, imo, which is the case, regardless of what speakers & or amps you are using.

i have two amps that can drive most any speakers, regardless of load, especially if they're being crossed over at ~80hz - a mesa baron, & a pair of audio mirror set mono blocks, that use a pair of 6c33c power tubes.  the mesa will make either 55, 85, 120, or 150wpc, depending on whether it's set at full triode, full pentode, or 1/3-2/3 triode/pentode in between.  the audio mirrors put out 40wpc of set.  i have a coupla other low power amps that work great on hi-efficiency & hi (or stable) impedence speakers, which are easier to drive.  my almarro a205a-mkll, at 5wpc, is incredible matched w/my coincident victory's rated at 97db/14 ohms, w/impedence range of 10-18 ohms.  my mesa drives my 91db/4 ohm piega p5mkll-ltds extremely well, from the 55w all-triode setting.  (i haven't tried the audio mitrrors w/these speakers, yet.)  the audio mirrors also drive the victory's great, but it's a bit of overkill in that application, imo.

again, none of these amps & speakers will ever see anything below 80hz or so, those duties being performed by a pair of solid-state electrocompaniet aw75dmb's driving vmps larger subs... the ec amps, btw, are known for being "tube-like", but these amps, once my main speaker amps, before being relegated to subwoofer use, (replacing a pair of series l adcom gfa555's), never got reinstalled w/my mains, once i got a taste of tube amps.  and, i cannot imagine ever going back to a solid state preamp, fwiw...

ymmv,

doug s.

Tubes for mid/high and SS for bass.   

Woodsyi and George:

The combination of SS for the bass and Tubes for the mid/Teeters would seem to be a good match.  I know that at least one person is going that route with his new Emerald Physics CS2 Speakers.  I believe the CS2s have an external crossover and you have to biamp (from comments I have read) but isn't it difficult to biamp most speakers?  I am just too new to this hobby to understand how to do so.

Ken

[Bob Reynolds]

Steve, please explain why tube amps use output transformers and why they typically have a few output taps? Intuitively, I've always assumed that the transformer was there to "match" the impedance of the amp to the loudspeaker. So, you'd use the 4 ohm tap when driving a "4 ohm" speaker. But, I really do not understand the benefits of matching the impedance.

Sure.

One reason output transformers are used is because generally a tube can't supply sufficient current into a low impedance load to get any appreciable power. That's largely due to the fact that tubes generally run at very high voltages; a couple hundred volts and more.

For example, to deliver 1 watt of power into 8 ohms, you need to deliver about 350mA of current. Let's say the tube's running at 300 volts. 350mA times 300 volts is over 100 watts.

But let's say the load is 2,000 ohms instead of 8 ohms. 1 watt of power into 2,000 ohms is only about 22mA and at 300 volts, you're only looking at about 6 watts.

So, how can we take that 1 watt we've got into the 2,000 ohm load and transfer it to our 8 ohm speaker load?

We can use a transformer.

To get that 1 watt into 2,000 ohms, the tube is swinging 44 volts. And in order to get that down to 2.83 volts for our 8 ohm load, we can use a transformer with a step-down ratio of about 16 to 1.

And while transformers will step voltages up or down as a function of their turns ratio, they reflect impedances by the square of their turns ratio. So for a transformer with a 16:1 voltage/turns ratio, impedances get reflected by a factor of 16 squared or about 256.

And if we multiply 8 ohms by 256, we get about... 2,000 ohms.

In a typical tube amp, say a single-ended triode amp, the tube's output (the plate in this case) is connected to one end of the output transformers primary and the other end of the transformer's primary is connected to the power supply (or B+). So the load the tube sees is really the loudspeaker's impedance as reflected to the transformer's primary, or in this example, 2,000 ohms.

The reason for the multiple taps on the transformer is because a given tube works most linearly into a given load. So in order for the tube to see the roughly same load whether you're using a 4, 8, or 16 ohm speaker etc., multiple taps are wound with different ratios.

For example, let's say the speaker load is 4 ohms instead of 8. To get the same 2,000 ohms reflected to the transformer's primary, we need a step-down ratio not of 16 to 1 as in the 8 ohm case, but about 22 to 1 which means the 4 ohm loudspeaker load is reflected back to the primary by a factor of 22 squared or about 500.

Of course this means that the 44 volts the tube is swinging into the 2,000 ohm primary load gets stepped down by a factor of 22, so instead of 2.83 volts across the loudspeaker, it's just 2 volts. And 2 volts into 4 ohms is... one watt. Same as for the 8 ohm tap with an 8 ohm speaker.

And that's why tube amps are typically rated for the same power into 4, 8, 16 ohms, etc.

This help?

se

Yes, thanks Steve. Very good explanation.

[JohnR]

Please refer to the posting guidelines, which say...

When using the "quote" button, please edit the quote to remove unnecessary text. In other words, try to avoid quoting 20 lines of text just to add one of your own.

http://www.audiocircle.com/index.php?topic=37305.0

Thanks

[Duke]

Steve,

Thanks for answering my question.  I won't take your time up with any more.

If anyone else would like to see some measurements, let me know.

Duke

[Audiovista]

If anyone else would like to see some measurements, let me know.

Duke

Duke - I would, always 

A good experiment is a great learning tool, even if you just read about it. 

....but please illustrate/describe your methodology when performing measurements....to paraphrase something I heard long time ago.....if you repeat measurements long enough, the results will tend to support theory you subscribe to.... You don't want that... 

There was a thread discussing series vs. parallel connection of speakers (in the line array arrangement, I think).

http://www.audiocircle.com/index.php?topic=45116.0

Best,
Boris

[Harry X]

Hi Guys, I am using both SS and tubes. I am using a Jolida 300b with my Tyler PD30's which sounds pretty impressive. The SS amps I tried with the Tyler's just didn't sound as good as the Jolida.  I also have a pair of Horns from the Horn Shoppe. I found that the Horns sounded good with both SS and Tubes. I have been switching back and fourth between a Sim Audio i5 LE int. and a Musical Fidelity a308 int. with the Horns. Both sound similar but I think the i5 is more refined, it operates in class A for the first 5 watts which is probably more than the horns are gonna use. Not sure what class the 308 runs. That being said I have read that the Horns work well with amps that have low dampening factor and High output impedance's which both the MF and Sim Audio do not have! I would love to try an F1 from First Watt, which has high output impedance and a low dampening factor to compare to see if there really is that much of a difference in sound.   Do these factors make that much of a difference?

[Duke]

To Boris (Audiovista) and Steve Eddy,

Well I made some measurements and to summarize it looks like Steve is right and my loudspeaker modelling program (which I'd assumed to be infallible) is wrong!  Misplaced trust in my quick-and-dirty modelling program had led me to some erroneous assumptions.  Let me explain:

Many moons ago when I first measured a difference in the bass region when going back and forth between a solid state amp and a high output impedance tube amp, I modelled the theoretical change in Qes on a loudspeaker modelling program and the results it gave me were very different from what I was observing.  So I was convinced that there was more to the picture than could be explained by an increase in Qes - it never occurred to me to question the modelling program.

The modelling program had predicted the increase in Qes would give me an average of about +1 dB from 60 Hz on down to 25 Hz, and I was measuring a difference that averaged closer to +4 dB over that region, with the measured increase greatest around the impedance peaks (which hadn't been predicted by the modelling program).  So I was convinced there was more to it than just the increase in Qes.

In this latest round of measurements, once again I found a significant difference in the bass region between the solid state amp and the high output impedance tube amp.  But this time I made an additional measurement - I inserted 7 ohms of series resistance on the solid state amp.  The result was not identical to what I was measuring on the OTL amp, but it was close enough that I could see that Steve was right and my loudspeaker modelling program was wrong.

So science has triumphed over Duke. 

I'll try to post some of my measurements, along with a description of how I made them.

Duke

[TerryO]

Duke,

This is pretty much what Nelson Pass has been saying for some time on the diyAudio forums regarding "Voltage vs Current source" amplification. The addition of series resistance after a SS amp, to lower the DF is an old trick to improve the perceived bass output of a speaker. I know that some speaker designers, like Greg Monfort (GM), usually ask if SS or Tubes will be used with a speaker design, and adjust their calculations accordingly. Obviously, for most mass market applications there are certain assumptions being made, that may not be warrented. 

Best Regards,
TerryO

[Bob Reynolds]

JohnR, I apologize about the long quote. I wasn't aware of the guidelines -- they make sense. Thanks for bringing it to my attention.

+++++++

On Bob Carver's Sunfire amps he has two pair of speaker binding posts: one for "current mode" and one for "voltage mode". The current mode pair has a 1 ohm resistor in series and he suggests that pair be used for the treble to give it a tube flavor. The normal voltage mode pair is for the bass for better control.

+++++++

I posted this in the lab a few months back. It didn't get much attention then. It's relevant to the current discussion.

http://stereophile.com/reference/810/

[Duke]

I hope these measurements are useful information to people interested in the topic of amplifier/speaker matching. 

First let me describe my test set-up.  I made all measurements using LinearX's LMS version 4.  The "wooferbox" is a 2.5 cubic foot test box tuned to 35 Hz, with a 12" prosound woofer in it.  I placed the wooferbox on the floor away from walls (that was convenient), and placed the microphone about 12 feet away and close to a corner (that was also convenient, and some corner reinforcement would make sure the data was well above the noise floor).  I did not use time-gating.

The room is my basement, and it's not ideal but hey it's paid for.  My goal was not to get nice pretty ad-copy-quality curves, but to be able to look for differences by comparing the curves.   Once the measurements began, I did not touch either the microphone or the wooferbox.

**********

Here's the impedance curve:



You can see the tall twin bass impedance peaks, and the inductive rise at high frequencies.  Also, looks like there's a resonance at 230 Hz.  That's not predicted by the internal dimensions, so I don't know what it is - but it doesn't matter for our purposes.

**********

Here we have the measured frequency response curves of the wooferbox driven by a high quality solid state amp and by a high quality OTL tube amp.  I matched up their levels in the region where the impedance is roughly 8 ohms, so it's easy to see the differences:



The red trace is the solid state amp, and the blue trace is the OTL tube amp (no political implications intended).  Room effects dominate below 200 Hz - in the real world, this is no surprise.  Anyway as you can see, below 100 Hz the OTL amp is generally about 3 or 4 dB up, with the biggest differences (over 6 dB around 25 Hz and 55 Hz) corresponding with the peaks on the impedance curve (scroll up a bit and you'll see).  From 100 Hz to 600 Hz the two curves are quite close together, though you can see that between 125 Hz and 250 Hz the solid state amp is up a bit due to the impedance dip in this region.  Finally above 600 Hz the curves diverge once again, as the OTL amp puts out increased power into the high impedance presented by the voice coil's inductive rise.

Just eyeballing these curves, I think you can see that the tonal balance would differ depending on which amp is used.  Of course adding a tweeter would make more of a difference than careful amplifier choice!

**********

I recall that the OTL amp's output impedance is in the 7 ohm ballpark, so I went back to the solid state amp and wired a 7 ohm resistor in series between the amp and the speaker.  I increased the gain so that the measurements came out very close to one another in the region where the speaker's impedance is around 8 ohms, as this makes it easier to see the differences:



The red trace is the solid state amp direct, and the green trace is the solid state amp + 7 ohms.  As you can see, adding 7 ohms of series resistance produces change identical in pattern to what we saw when comparing to the OTL amp.  Steve Eddy, you were right!  I WAS making things much more complicated than they really are.

**********

Before we go, let's compare the OTL amp with the solid state amp + 7 ohms series resistance:



The blue trace is the OTL amp and the green trace is the solid state amp + 7 ohms.  It looks like the actual output impedance of the OTL amp may be a bit higher than 7 ohms.

I hope this shows why amplifier-to-speaker matching can make a difference if you're looking at specialty tube amps.  In particular, I hope it shows why the bass response of some speakers may either benefit from, or be degraded by, an amplifier with a high output impedance.   

**********

Finally, a speaker designer who intends to offer speakers compatible with high-output-impedance tube amps might want to keep the impedance curve as smooth as possible above the bass impedance peaks so there's no significant change in tonal balance from one amp to another, and perhaps also tune the speaker a bit lower than normal (or offer that as an option) to take advantage of such amplifiers' increased power output in the deep bass region.  Here's what the impedance curve of such a speaker might look like:



**********

I took another set of measurements as well, using cone midrange drivers wired to present different impedance loads, comparing how these two amplifier types behave into different fairly smooth loads (4, 8, and 16 ohms).  If anyone would like to see those, let me know.

Duke

[Duke]

Hi TerryO,

I tried using series resistance as part of a crossover once, hoping to make the speaker an easier load while warming up the bass a skoshe.  Unfortunately I didn't take any measurements that would back up this subjective assessment, but the series resistance (only 1 ohm, high quality and very high wattage) seemed to suck the life out of the sound.  Haven't tried it since.

Nor have I played with a Sunfire amp, but my guess is that one ohm of series resistance could either cause the frequency response curve of the speaker to zig or to zag a bit, depending on the impedance curve.  But I don't think it would make a dramatic difference with most speakers.

My impression is that while it's possible to a certain extent to emulate the characteristics of a tube amp by using series resistance on a solid state amp, I'm not sure the net sonic result is going to be in the same league.  But maybe with a high enough quality resistor, it would be.

Duke

[TerryO]

Hi TerryO,

I tried using series resistance as part of a crossover once, hoping to make the speaker an easier load while warming up the bass a skoshe.  Unfortunately I didn't take any measurements that would back up this subjective assessment, but the series resistance (only 1 ohm, high quality and very high wattage) seemed to suck the life out of the sound.  Haven't tried it since.

Nor have I played with a Sunfire amp, but my guess is that one ohm of series resistance could either cause the frequency response curve of the speaker to zig or to zag a bit, depending on the impedance curve.  But I don't think it would make a dramatic difference with most speakers.

My impression is that while it's possible to a certain extent to emulate the characteristics of a tube amp by using series resistance on a solid state amp, I'm not sure the net sonic result is going to be in the same league.  But maybe with a high enough quality resistor, it would be.

Duke

Duke,

Two years ago, I built a 2-way speaker designed by a friend that utilized a series crossover (2nd order). When I was wiring up the crossover I left out a resistor that he had indicated as an option. I had incorrectly assumed that it was there if the speaker was to used with a SS amp. The event I built it for is well known to be almost exclusively using tubes (lots of SETs) so I left it out. After the event I spoke to him and commented on this. He told me that he had the optional resistor there to pad down the tweeter a few dBs. Well, he was intrigued with the notion and started playing around with different values of series resistance. I believe he ended up with somewhere between 3 and 4 ohms and the bass shelf was extended a fair amount, at least with a SS amp. IIRC, the treble didn't suffer a bit, and that seems in keeping with Nelson Pass's opinion that the series resistance can actually improve the treble by "swamping out" the impedance rise in the upper frequencies (Note: He's actually addressing "fullrange drivers" in this context, but it also seems to hold true in this instance).

Best Regards,
TerryO

[Duke]

Thanks for the additional information, TerryO.

Maybe I cut and ran on the idea too soon - I have found a better high power resistor since then, and maybe ought to consider it should the occasion arise again.

Duke

[TerryO]

Thanks for the additional information, TerryO.

Maybe I cut and ran on the idea too soon - I have found a better high power resistor since then, and maybe ought to consider it should the occasion arise again.

Duke

Duke,

It's late and I'm a little fuzzy, but I think that the nearly ten fold increase in the impedance of your test driver may be masking the differences to an extent. I know that it just drives most tube amps crazy (Booming, Bloated and flabby Bass) when they encounter big impedance swings like that. 

Best Regards,
TerryO

[Duke]

TerryO,

I'm not sure what you mean by the tenfold increase in the impedance masking the differences between the amplifiers.  I think it throws a spotlight on them.

Regarding your comment that an impedance peak in the bass region "drives most tube amps crazy" - if the speaker was designed for a voltage source [solid state] amp, in many cases the tuning frequency may well be be too high for the application - resulting in the bass being too loud near the tuning frequency, which is what we interpret as "bloat" or "boom" or "flab".  This is actually a speaker/amp interaction problem, not the amp being driven "crazy" - but it can sound like that's what's happening because it appears when the tube amp is put in the system and disappears when it is removed.

I don't think my data shows an amp being driven crazy.

If the enclosure is tuned taking the amp's characteristics into account, which if we're talking about a high output impedance amp means tuned considerably lower than normal, then the bass ends up nicely balanced.  I forgot to do a measurement of my test wooferbox with a longer port before I took down my test setup, and that would have shed light on this issue.  Lowering the tuning frequency moves the frequencies that are boosted downward, so that instead of "bloat" you have deeper extension.  You can extend the bass as much as 1/3 octave deeper with this technique.  I can post links to online commentary that indicates it's subjectively "tight" as well.

Room acoustics (including positioning of speakers and listener within the room) also play a large role in the bass region (in fact I would say that room acoustics dominates, but that's another topic).  I don't know why more manufacturers don't offer variable-tuning bass systems, as the odds really are against the factory setting being what works best in a given room.

Duke

[Wind Chaser]

This may be a bit premature, but it's beginning to look like the conclusion of the whole matter of isn't one of absolute design topology independent of the burden it bears...

Recognizing that, Paul Speltz (a music loving Electronic Engineer) made a name for himself with his ZERO Impedance Autoformers. - click here for more info - Initially designed for OTL amps, people have found them to be of great benefit with virtually any kind of design including SS.  Myself and others have used them in "reverse" to lower the impedance with very good results as well.


John

[dado5]

I was much like Wayner - preferring tube preamp driving SS amps - the 3 dimensional nature was good.  But 10 seconds in to hearing a SS pre-amp driving mono tube amps and I was sold on that combo instantly.

In my case I went from a Dynaco PAS-4 (tube preamp) happily passing it to Odyssey Khartago Extreme amps to something I've even happier about now.  It has nothing to do with Odyssey offerings - which are superb SS amps (the best SS amps that I ever owned)

The lowest level signals, especially if you enjoy vinyl (typically, 1/20th to 1/400th or less voltage than CD outputs) beg for as clean a step-up as possible.  If you pass that to a capable tube amp(s) (that is, one that is capable of driving your speakers with adequate headroom for peak passages) you have ultra-quiet step-up that can only be achieved with solid state with the purity and  3 dimensionality that tubes bring to your music.

SS preamps seem to provide a 'backbone' of sorts to the music that I haven't been able to approximate with tube preamps.....even those that have been solid state regulated.

If you have tube amps that are either tube or solid state regulated (not rectified), you achieve considerably more chest-thumping bass 'tightness' that is usually the domain of solid state amps.

If you add a powered subwoofer(s) to the mix - you're further relieving your tube amp of the heaviest bass duties and you can achieve some smile-inducing bass heft and 'tightness' back into the sonic realm 

So, SS preamp and tube amps for me - an unlikely electrical match as tubes typically step-up voltages better and solid state better develops current (all the better to drive most complex speaker loads).  The more you simplify your speaker loads, the more tolerant of tube amps your system will be - and you can enjoy the excellent pairing of a SS pre-amp and tube amp(s) more completely

If you're into vinyl especially, you might want to turn your Odyssey / Candela up on it's conventional head and try it...I've never been happier 

My current system:

Bell Dual Mono Integrateds (10 watt tube units powering 6.5" mid-woofers and tweeter)
Mitubishi DA-C20 working as SS phono preamp and providing AM/FM
Gallo TR-1 (100 watt) subwoofer (dedicated below 80hz)
Linaeum Tower speakers (80hz and up)

CD is run straight into the Bell's....the added liquidity of all tube seems to benefit CD playback.

 

That's an interesting take.  Have you seen Jack Eliano's new design?  OpAmp voltage gain transformer coupled to a 300b output stage, single ended and push pull.
]http://www.electra-print.com/singleended_a2.php[url]
http://www.electra-print.com/pushpull_a2.php[/url]