Where does that leave the tube preamp and SS power amp brigade? A bit of added 2nd harmonic?
Regards
M. Gregg
I think tube preamps just generally do a better job of amplifying the signal cleanly than the typical Opamp or multistage SS preamps. People like them because they can be intrinsically better. My preamps have no valve euphonics what so ever.
Shoog
I agree,
(up to a level of gain).
Then again there is Mr Miller! and the triode pentode debate..
Regards
M. Gregg
Looks like its back to,
the 50's..that amp has nice tone...
A 211SE is a very expensive way of getting it!
Regards
M. Gregg
the 50's..that amp has nice tone...
A 211SE is a very expensive way of getting it!
Regards
M. Gregg
Last edited:
Have you ever done a level matched unbiased and aural memory span compensated listening comparison of amps?
+1
I wasn't speaking of the OP's question, but your question:20to20 said:
If I had known that you had the graphs in front of you I would not have needed to estimate an answer to your question. Out of interest, how does my estimate compare wkth the graphs?20to20 said:
Have you - do you base your opinion on such tests ? Or do you build amps - put them in your system for a few weeks/months - swap them out for another amp and then draw general conclusions about how they sound in the real world. Thats the way I do it and its generally fairly good at finding even subtle differences between amps (even one of almost the same design).
In the end I have found that mostly the differences are insignificant and vastly overwhelmed by the similarities. Then I have to ask whats more cost effective, which one behaves better if I have to push it - which is more versatile with a range of preamps and speakers.
However - I always study the subject as broadly as possible and use the results (both measured and subjective) of others who have proven their designs and incorporate them into my own designs. I become very suspicious when distortion figures are very high and the dominant tone depends on the intrinsic limitations of a single component (OT mainly).
Shoog
Last edited:
http://tdsl.duncanamps.com/link.php?target=0925A019
7189-A data sheet I refered to. Pdf
See if you can interpolate the PP distortion graph, leaving the operating point set, but changing the input signal to stay out of cutoff. What power level is that? Then go to the SE chart and find the THD for that power level.
But post #1 is still a much simpler question that should have been addressed before all "camp supporting" posts came rolling in.
That's interesting..
So are we saying a class B PP amp is "Better" THD than a class A SE..
Humm...I must reflect for a while..🙂... 😀
Regards
M. Gregg
So are we saying a class B PP amp is "Better" THD than a class A SE..
Humm...I must reflect for a while..🙂... 😀
Regards
M. Gregg
Last edited:
I'm not sure that is an interesting question to ask, and the data to answer it is not provided. What can be done is to compare THD PP with SE at the same power level.20to20 said:
0.5W 1.6% SE, 0.8% PP
1W 2.8% SE, 1.0% PP
2W 4.4% SE, 1.2% PP
4W 6.4% SE, 1.7% PP
Yes.
If you do a listening comparison of 2 of the same model amps playing at different levels, you will hear the difference. Think about that.
Next time you want to debate about amp sound, make sure you do a level matched unbiased and aural memory span compensated listening comparison first. Try it, it's an eye opener!
Those numbers look like the SE has lots of 3rd and the PP is mostly 2nd,
so what's wrong with this picture?
😎
If the first question and the following sentence are combined as an idea, SE distortion vis-a-vis PP, and answered, then the following ones wouldn't need to generate 300 meandering responses bounced off each other.
Something I've wondered about when it comes to data sheet distortion specs, SE vs. PP, is that the manufacturer never specify much about the quality of the OPT's other that the Z. They never state where the test signal is taken from, is it always the plate? Is it always the secondary?
If it's the plate then there would be no cancellation to consider. If it's the secondary, and cancellation is factored in, then the distortion is still a product of the tube whether it's an SE or PP. Unless there is some other inhearent benefit of the primary winding circuit of a PP output that contributes to lower distortion products from the tube. Never heard of any such explanation except the benefit of a UL tap.
20
It's the opposite of what might be expected. Maybe the PP figures are mainly distortion from the driver/PS, and the PP output is largely linear?Nelson Pass said:
You can assume that good quality transformers will be used. However, it is possible that sometimes they cheat and put a big resistor across the primary instead of a small resistor across the secondary. I don't know.20to20 said:
Not true. If there is a PP OPT, or just tightly coupled primaries with no secondary, then cancellation takes place.
I don't understand. Distortion mainly comes from the valve, whether SE or PP. Its just that for PP the even-order distortion, although generated in quite large amounts, is cancelled so doesn't appear in the output.
It depends a lot on the load! The distortion spectrum for a SE pentode changes a quite a lot with the typical load provided by an average speaker from about half of its max power output. Below it is more or less the same. It is one reason why pentodes when run in SE need feedback regardless of damping factor. The EL84 SE at 250V anode voltage with 5K load has dominant second harmonic up to the onset of clipping. In fact crossover between 2nd and 3rd happens around 5W output. If the load is 7K or 3K crossover happens just above 3W. Also the total amount is affected. There is a kind of parabolic dependence of THD as function of the load.
Triodes are much better in this respect, once one has the right device for his target Pout there is no surprise...rather lots of people insist with low powered amps that inevitably have large amounts of THD at listening levels where it shouldn't be...
For the PP, 3rd is usually the main, and often the only, harmonic up to good level. Dominant 2nd harmonic can happen for certain amounts of Ultralinear connections such as 20% in the case of EL84. However this is also the UL configuration that gives almost the the same efficiency of the pentode but more than 3 times lower distortion at any power level up to 15W. On the other side it has the same distortion of the "classic" 43% UL connection but 50% more power! The fact that 2nd is dominant for the 20% UL might not be (it is certainly not for me) so relevant as THD is only 1% a 15W without any feedback.
THD figures in datasheets assume the rest of the amplifier is ideal.
Even if I grant your point, I see this tendency even at the low end of the
wattage in the data.
Something else is going on, and I will submit that it makes this particular
case suspect as a good example of whatever point is being made.
When I think about the subject, 2nd harmonic dominance comes to mind when
the stage is SE Class A, and 3rd harmonic for PP. I can't help but imagine
that there is a general consensus on this, and the question mostly revolves
around the trade off between the higher 2nd harmonic of SE and the lower
THD total concentrated in 3rd.
Of course we understand that the typical example of SE goes toward 3rd
toward the top end of the power curve - as you say, somewhere usually above
the halfway point.
😎
Yes the tendency is there but the amount and balance of harmonics doesn't change a lot at low power.
It happens with all power pentodes I know with a gentle knee as this was considered desirable for HiFi. So it is possibly a linearity issue. The discriminant component for the end result is the 2nd that can be nearly zero with the optimal load at full power or the dominant one for lower or higher anode loads. The third shows an increasing monotonic behaviour both as function of the load and Pout for given a load.
I prefer low THD first and with complex signals 2nd harmonic might not be so pleasant to listen to as intermodulation products might have nothing to do with the harmony of the originals. I always try to cancel as much as possible 2nd harmonic in my SE designs before using feedback so that 2nd harmonic doesn't grow beyond a certain level. I get the crossover between 2nd and 3rd well before the clipping also with triodes, with lower total amount at all levels compared to a conventional design using the same devices. Of course Pout is higher as well for a given THD, in particular it's a lot higher if the driver can cope with some class A2 or AB2 operation.
It happens with all power pentodes I know with a gentle knee as this was considered desirable for HiFi. So it is possibly a linearity issue. The discriminant component for the end result is the 2nd that can be nearly zero with the optimal load at full power or the dominant one for lower or higher anode loads. The third shows an increasing monotonic behaviour both as function of the load and Pout for given a load.
I prefer low THD first and with complex signals 2nd harmonic might not be so pleasant to listen to as intermodulation products might have nothing to do with the harmony of the originals. I always try to cancel as much as possible 2nd harmonic in my SE designs before using feedback so that 2nd harmonic doesn't grow beyond a certain level. I get the crossover between 2nd and 3rd well before the clipping also with triodes, with lower total amount at all levels compared to a conventional design using the same devices. Of course Pout is higher as well for a given THD, in particular it's a lot higher if the driver can cope with some class A2 or AB2 operation.
Last edited:
What the measured data presented in this thread points to is that PP amps behave in a far from ideal way even at low levels. In this case tube imbalance makes the second harmonic the dominant distortion above third in the Class A amps shown. As has been shown for the specific amps discussed SE and PP are not so different in this respect.
Shoog
Shoog
- Status
- Not open for further replies.