Well you are driving the ESLs with the PS.... if DeWald.....has a SE output tran ...that well sound diff......... Glad it works ...an sounds good... in the end the all the matters...have fun
http://education.lenardaudio.com/en/14_valve_amps_4.html
http://education.lenardaudio.com/en/14_valve_amps_4.html
Last edited:
The only reason to recommend lowering the NF would be if adding the capacitive load of your ESL was causing instability and/or HF oscillation. If this is not the case, then there would be no benefit to lowering the NF, that would offset the detriments you mentioned.
Yes, differences in tube characteristics will show up when driving ESLs directly from the plates. These differences are mainly due to the interaction between the tube internal impedance (operating voltage vs current transfer) characteristic and the load impedance it sees. Basically when connecting to the plates, you are picking off output from a voltage divider made up of the tube internal impedance and the load impedance. The load impedance is a combination of the transformer primary inductance, core losses, and reflected load from the secondary. So, you may also notice a change in sound quality by changing the value of the load resistor you are using, or which tap it is connected to.
Note that the impedance of loudspeakers are not constant like your load resistor, but varies with frequency both in magnitude and phase. So while you may notice changes in sound quality when switching tubes while listening to your ESL hooked directly to the plates, it may not be the same quality change you would experience if driving loudspeakers from the transformer secondary.
Thanks bolserst for the reply. Interesting about changing the load resistor. I'm using an 8 ohm load resistor but I'm thinking on trying a 16 ohms on the Fisher. The DeWald actually has a 32 ohms tap.... I may try that too. Probably a much easier load for the tubes.
How will the quality of the output transformer affect the sound on the ESL. In other words, would a better transformer cause a more linear response on the ESL?
Thanks.
How will the quality of the output transformer affect the sound on the ESL. In other words, would a better transformer cause a more linear response on the ESL?
Thanks.
Last edited:
I hope this reply won't spoil the fun too much, but... The transformer is part of your feedback network, so even with an ideal infinite-gain amplifier, the overall sound quality will directly depend on the performance of the transformer. If you want to change that, you have to somehow take the feedback directly from both plates, but I'm not sure how to do that in an otherwise unbalanced amplifier.
As MarcelvdG mentioned, the transformer is part of your feedback network so there is the possibility that voltage on the secondary differs slightly from voltage at the plates. However, for frequencies below where the leakage inductance becomes dominant, the feedback loop will be driving the plate voltages to exactly follow the input voltage. Without getting too technical, below about 10kHz(where leakage inductance often starts affects response) the voltage wave form on the secondary will follow exactly the plate voltages applied to the primary because they are linked to the same magnetic flux changes which are all in the core. Above 10kHz there becomes progressively more and more magnetic flux generated by the primary that are not linked thru the core to the secondary. This manifests as an inductance in series with the secondary(the so called leakage inductance). At higher frequencies, the feedback loop will be driving the plate voltages to NOT follow the input voltage to make up for the impact of the leakage inductance and get the secondary voltage to follow the input voltage.
So, a better transformer will have lower leakage inductance and extend the range over which the secondary voltage follows the voltage applied from the plates to the primary. It is easier to make low-power transformers with low leakage inductance than it is hi-power transformers. This may be part of the reason you are preferring the sound of the lower power DeWald amp.
This can be done if the resistive load is moved from the secondary to the primary side. You would then just be using the primary side of the transformer as a load inductor. You would need to include an appropriately sized voltage divider to use in applying the feedback. The feedback loop would no longer have to correct for the effect of leakage inductance. It would still, however, have to correct for changes in the load due to core related non-linearities. To remove this source of distortion you would need to remove the transformer and replace their load inductance with load resistors which would generate lots of heat and require double the supply voltage. (ie see Sanders Amp)
Last edited:
You could take the feedback from one of the plates, but then you would still be using the transformer primary as an autotransformer. I'm not sure if that is any better than using it as a normal transformer. It is not obvious (not to me, anyway) how to take the feedback signal from both plates, as the signal on one of the plates has the wrong polarity. You haven't got this issue in a fully balanced amplifier.
It is better because you have removed the leakage inductance and its interaction with the load impedance from the feedback loop so the plate voltage waveform more closely follow the input voltage at higher frequencies(ie > 10Khz).
To use both plates you would need to have an amplifier with a differential input stage like the Sanders amp, or use a differential amplifier(long tail pair in tube-terms) to sense the current thru a portion of a load resistor connected between the plates and then inject this signal at the cathode resistor of the input stage as is typically done.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.