Primary impedance for 4x6L6?

[Hallingamp]

Hi all.

This is probably a simple question, but I'd like to be sure before I start spending money...

Do you think a 4000 ohm primary OT is too high for 4x6L6?

Operating conditions will be 510V on the plates, and cathode biased for 23-24 mA/tube. And I'm planning to use the UL taps. Haven't tried that before.

Am I right in thinking the 6L6 is a "potato"-tube that will just say "Oh OK, close enough."?

Thanks in advance.

Jan

[Rob Mercure]

There are different impedances for differing desired operating conditions and those that deliver the most power aren't necessarily those that produce the least distortion - for example. For guitar amps power is usually a primary consideration and the most common impedance for a push pull pair of 6L6 tubes is 6.6K. Putting two tubes in parallel halves the impedance so a 3.3K primary would commonly used - 4K ain't that far off so you'll probably be fine with it.

Rob

[Satamax]

http://tdsl.duncanamps.com/show.php?des=6L6

[Gregg]

If you're building a guitar amp most of the time the OT primary for a 100 watter /4x6L6 or 4xEL34/ is between 1700 and 2000 Ohms. Check out for example Fender OTs specs - most of the time it's a 4-4.2k for 2x6L6 and 2k for 4x6L6.

[Le Basseur]

(...) Check out for example Fender OTs specs - most of the time it's a 4-4.2k for 2x6L6 and 2k for 4x6L6.

+1
I'd not steal or divert the initial thread,but can anyone explain the under-valued Fender OT primaries?...aside the fact that an output tube is a VERY forgiving/tolerant piece of equipment,as opposed to most SS.
When Fender mounted those 2k in a 4x6L6 Bassman head,it was an engineer's decision or an economical one?
(I really don't want to open a can'o'worms but...)
There was a famous issue a few years ago when Mr.Nobu Shishido (a well-known and respected tube hi-end guru) published his well-documented analysis on the different OT primary impedances in a SE 300B tube amp.The conclusions were relevant:all the other things/parametres being equal,a higher OT primary implies a significant reduction of the harmonic distortions,IIRC up to 15% or so,on the expense of a slight output power loss.
Could be the Fender's smaller OT primary's value the result of of a "let's-get-more-harmonic-distortions-for-a-more-specific-sound-and-more-power" kind of decison?
Any comment and correction greatly appreciated!
Thanks in advance!
Regards,

[bob p]

i've also been scratching my head about the OT primary impedances for multiple tube arrangements. in my case, i was looking at Hammond iron for high powered HiFi amps. in all of the cases, the OT primary impedance for single pair PP topologies were right where you expected them to be, and the OT primary impedance for double pair PP topologies were WAY lower than I expected them to be. i'm at a loss to explain it.

[Gregg]

Could be the Fender's smaller OT primary's value the result of of a "let's-get-more-harmonic-distortions-for-a-more-specific-sound-and-more-power" kind of decison?

I suspect this to be the reason.

[AdmiralB]

It's because guitar speakers aren't really anywhere close to their rated impedance for most of the frequency range a guitar generates.

Go to the Eminence site - they have impedance curves for their speakers. Look at a 16-ohm model - you'll see that the only place it's anywhere close to 16 ohms is between about 300Hz-450Hz. The rest of the range is more like 25 ohms. 16/25 = ~64%, which (not) coincidentally is the about the same ratio as a 4K OT to a 6.5K OT.

Hi-fi is different, because the amp is typically driving a crossover than tends to normalize the load.

The Marshall VM has "hifi" OT specs, too. The 50W is 6.7K, and the 100W is 4K. Both make more power (and live happier lives) mismatched down.

[bob p]

Go to the Eminence site - ...

i am familiar with the notion that the impedance of a speaker varies as a function of frequency. as it turns out, i have even used my test gear to generate the types of plots that you are referring to.

Hi-fi is different, because the amp is typically driving a crossover than tends to normalize the load.

yes, but...

i'm still missing something here. BOTH guitar amps and HiFi amps have OTs whose primary Z is lower than expected, regardless of whether they drive a crossover network or not; in both cases, the primary Z for a parallel PP configuration is significantly less than half of the primary Z for a single PP configuration. the statement that HiFi amps are different because they feed a "normalized load" doesn't explain why BOTH HiFi amps and guitar amps have a lower than expected primary Z for parallel PP configurations.

anybody got math?

[AdmiralB]

iin both cases, the primary Z for a parallel PP configuration is significantly less than half of the primary Z for a single PP configuration.

I think this is where you're going wrong. The primary Z for a pair of 6L6s in 99% of guitar amps is in the 4K-4.5K range, and the Z for a quad is half that. IME it is ONLY amps that use OTs originally intended for other applications - notably the JTM 45 - that you see the OTs with 6K and higher primaries.

[R.G.]

anybody got math?

Nope. Got charts.

I have some pics from 50's hifi mags somewhere that shows plots of power out, second harmonic and third harmonic as a function of plate to plate loading for a pair of pentodes.

It turns out that all of these curves have different shapes. The typical place where the second and third harmonic distortion is low and where the power output is fairly high is where they ran pentodes during the Golden Age. As I remember, this is about 6.6K for 6L6's, and is usually different per tube. The power output peaks at a different place, usually lower than the place where the second-third-power compromise is the best. For 6L6's it's around 4K to 4.4K, and is a broad, ill defined hump.

I don't know if there is any more math behind it than that. The curves obviously reflect the internal structure of the power tubes and their ability to put electrons over the lines cleanly, but I have never seen any credible sendup on how to calculate these curves a priori.

Engineering in the 50's was often graph-driven.

[bob p]

D'Oh!

the last two posts have made it clear to me that i wasn't thinking about the two roles of optimizing for power output vs. optimizing for low distortion. it makes sense that with a single pair of PP tubes, the OT could be optimized for low distortion (HiFi) or high power output (guitar amp), while the OT primaries for multiple pairs of PP tubes (HiFi) would be biased toward optimization for low distortion rather than power output, as the assumption is that power has already been achieved by adding the parallel PP pair. <slaps forehead>

Nope. Got charts.

I have some pics from 50's hifi mags somewhere that shows plots of power out, second harmonic and third harmonic as a function of plate to plate loading for a pair of pentodes.

the GE 6L6 Data Sheet is a widely available document that actually has a pretty good graph of what you're describing on Page 9. looking at the sheet, its easy to follow RG's comments.

its harder to find a good source for the charts that break THD down into its individual components. as it turns out I've got a good example of what RG's describing in one of my old electronics textbooks. its from the P-P & Phase Inversion chapter, and the graph was used to explain the difference in THD profiles between SE and PP amps. I've attached a chart below that's worth a kiloword.

RG, if you're still reading, could you take a look at this? Thanks!

[R.G.]

That is the kind of charts I meant, all right. I have only see the chart showthe power max at a lower load resistance than the two dips in 2nd and 3rd harmonic, but it does go to show that it varies with tube.

I will go reply on the diode thread about rectifiers.

[Bruce / Mission Amps]

I wonder how that follows with beam forming power tubes?
Regardless... the differences don't look as big as I thought it would be for the zed going up or down with power output.
For music production (not reproduction) is there anything wrong with a little more 2nd harmonic content?

[AdmiralB]

How much 2nd order is there going to be in a push-pull amp, regardless of Z?