A while back I posted comments about how great my new Shuguang EL-34B tubes sounded as compared to the regular El-34
(I'm using an ASL Aq1003dt amp that uses ultra-linear mode.)

I didn't understand WHY they sound great .. but the sound is just so much more "alive and rich"

I have also recently installed four EL-34 tube from EI in Serbia? .. but they are marked made in China.
So who knows.. I got them from MCM electronics.. but they didn't have any knowledge about them at all... lol

I wasn't too sure about these at first ... so I set them aside for a while.
However given more recent listening time I must admit that these (made-in-China EL-34) DO sound pretty darn good I have to admit after all.
BUT BUT ... but still... each time I swap the EL-34Bs back in the amp ....the difference is so nicely noticeable.

I don't know a lot about all the subtle difference in tube design or amp classes as it relates to sound differences... so I looked up what I could this week.

I finally may have discovered a hint as to WHY there is a noticeable difference.. but I don't completely understand the physics exactly.

The Shuguang manufacturer states:
"Plate power dissipation of EL34-B can reach 30W, whereas EL34A is 25W. EL34-B belongs to tetrode
tube compared to pentode tude of EL34A. g1 and g2 of EL34-B is not exactly aligned opposite to each
other and it is used in linear power amplification."

Then I read that "the beam-tetrode produces significantly less third-order harmonics compared to the pentode"

So I wonder if any of you here can expound on all this and tell me also if any of you have ever encountered and detected this difference in Tetrode vs Pentode sound.

BTW... I have four RCA 6ca7 tubes coming and I'm anxious to try them as well.
I cannot determine for sure if the 6ca7 is in fact also a beam-tetrode too.. or if it has the Pentode suppressor grid.
Does anyone know for certain?

I'm not surprised that MCM doesn't know anything about the tubes; that company has really gone down hill in the last couple of years. Most of what they sell is Chinese made junk. A couple of weeks ago I tried to place an order with them; I needed about 9 items. After hearing 'not in stock--back-ordered' , on the first 6 items, I decided to go to Digikey.

6CA7/EL34 is a power pentode. Loads of information on net. The Shuguang Works put out a range of products at various price points. They've been at it for a very long time, and if you are patient, you can have your questions answered.

I alway prefered the sound of 6CA7's over true EL34's . The EL34 is a true pentode , the 6CA7 and 6L6 is not they are more lkie the KT 66 .

Time to get out the oscilloscope. You can use a computer sound card to build a decent distortion analyzer too.

Hi Mescalero:
... Care to expound on that...lol
I'm all ears.

Also..In my original post I included a quote I found:
"the beam-tetrode produces significantly less third-order harmonics compared to the pentode"

I was hoping, with that quote, to awaken the minds of anyone here who more fully understands that "odd harmonic" aspect ... and might be able to confirm if this fact is that which is most significantly contributing to my sense that it "seems to sound better" ... and maybe that person could perhaps explain why that is ...if it 'truly IS the primary reason.

If you go back and read the RCA "white papers", beginning circa about 1936 or so, regarding the original development of the first beam-power tubes (the 6L6, essentially), you will notice that reduction of odd-order harmonics (especially third order) was a goal of the process, compared to the Philips pentodes (the main competition).

The rationale was- the 6L6 (and most other beam power tubes) were made for push-pull operation, primarily. In this operational mode, even-order harmonics are almost completely cancelled out, by the symmetrical nature of the two tubes working in see-saw with each other. In other words- even-order harmonics are usually caused by asymmetry between the two halves of a waveform- therefore push-pull amps, when calibrated for even gain on each side of the waveform (i.e, the tubes are matched, and the input to the tubes also matched so that each is doing the same thing), have very little even-order harmonic distortion. So, the manufacturer is free to concentrate more on minimizing odd-order harmonics... which they did, in the basic design of the beam-power tube.

Pretty clever, IMHO. And it seems to work. Or, at least my ears like 'em. If I had to pick a favorite power tube, the 6L6/KT66 would probably be it, with the KT88 second, and the 6CA7 third...

Regards,
Gordon.

Thanks Gordon... that's interesting helpful info. Exactly what I needed.
... now to further clarify the question...
When I pull out an EL-34 (identified as a pentode) .. and put in a EL-34B (which claims to be a tetrode)
.. Then I seem to hear a wonderful improvement....
is that improvement due to the harmonic reduction differences of the EL-34B ... or what else is it that makes such a difference?
.... "to me at least"

Even order harmonics are not unpleasant to the ear, but odd ones seem discordant and irritating. And there are even and odd harmonics in each and every stage of any amplifier. That is why terms such as THD (Total Harmonic Distortion) should be taken with a grain of salt if your doctor allows it.

When we get to the weakest link (in my opiion,) the loudspeaker, the odd order harmonics can cause displacement that may not be directly heard, but manifest themselves as unpleasant. It is also a mistake to disregard harmonics beyond the audible range unless steps are taken to take them out.

I agree that the weakest link are the speakers used. One could have a multi thousand dollar audio system, connect it to cheap speakers and get horrible sound.

See here:
http://www.qsl.net/dl4yhf/spectra1.html

Very cool what people are doing with this. The VLF direction finder app. is crazy cool!

Edit - more here:
http://g4oep.atspace.com/distortion/distortion.htm

Crickets in background..
I read the heck out of the second link. I liked the way he even explained how he used "Maple" to create the curves and transfer function graphs. Those of us with a scope have seen the crossover distortion before. Now, I didn't finish high school. So, don't be entirely discouraged. Just read through some of it. It can give a good mental image of what you're looking for. Then, when you see it, you'll know what it is. When you know what it is you have a good chance at solving it.

Seriously, no comments? I'm crushed!

OK, some examples:

First up is my E-Mu USB sound "card" driving into a (long) loop-back cable. Notice that it is (deliberately) over-driven. See the odd-order harmonics? These are typical of hard (i.e.: transistor-style) clipping.

Then, I turn it down to a more reasonable level. The SINAD button shows 67.9 dB of "SIgnal over Noise And Distortion." That equates to about 0.05% distortion and noise relative to the fundamental of 1 KHz. If you look down the "waterfall" display you may see where I set the sound card's signal level below clipping.

Here is a typical view of my work computer. I am capturing a schematic while watching some music show off its frequency/amplitude display on the side of my second screen. It's amazing how much high-frequency information is chopped off in some MP3 files. This one's actually pretty good. I think that this is Pink Floyd. Groovy..


So, you can monitor an amplifier with a sound card and free software. You can evaluate amps for distortion and IMD while tweaking bias, swapping tubes, changing feedback arrangements, whatever. I terminate the amp with a resistor, or in some cases speakers, and use a pot to tap off a little bit for my sound card. The rest is just whatever my little heart desires!

Comments?

Can you tell us more about the this software, I am very interested. Will it measure frequency response of a audio amplifier. How would you connect the output of the amplifier back to the computer to measure its response. Will it measure distortion also?

Thanks,
Sal

Mescalero,

Nice images, and good to see others using SC audio analyzer software.

Sal,

I use ARTA software http://www.artalabs.hr/ and a M-audio mobile pre usb soundcard.

With ARTA you can not only measure the response, noise and distortion of an amp, you can measure speaker impedance, and parameters and a bunch of other stuff.

Here's a website with how to use ARTA http://soundhobby.com/measurementsystem.htm

Thanks, Chuck.
Sal, I must warn that Spectrum Lab is a very low-level program. This both allows one to have incredible control over what it is used for and demands that one be at least reasonably proficient with audio and audio software/hardware. The program that Chuck recommended is probably a far better place to start. There are a lot of audio analysis software programs and many are free.
The sky's the limit!

Thanks for the information. I'll have to download some software and try them out.

Sal

That's the spirit! It's learning new things that keeps me truckin' along.

Sal,

I made a box similar to this one posted on the ARTA site http://www.artalabs.hr/AppNotes/AP1_Mea ... ev4Eng.pdf


I added an op-amp buffer to drive the voltage divider (R3, R4) to enable measurement of relatively high impedances. For low impedances of speakers the op-amp isn't needed.