@ cerem
with BF genrator : 1Vac on the secondary,and voltage read on the primary 🙂
or with variac : 100Vac on the primary,and voltage read on the secondary 🙂
@wavebourn
I have never seen such a way to do with the windings of an opt, thank you.
I have a clone of two leak TL25 + in a single chassis at home with custom opt that have special windings too 🙂
@ Waltube
vrey very interesting thread ,thank you !
with BF genrator : 1Vac on the secondary,and voltage read on the primary 🙂
or with variac : 100Vac on the primary,and voltage read on the secondary 🙂
@wavebourn
I have never seen such a way to do with the windings of an opt, thank you.
I have a clone of two leak TL25 + in a single chassis at home with custom opt that have special windings too 🙂
@ Waltube
vrey very interesting thread ,thank you !
Last edited:
> the "true" or "good" way to measure the reflective impedance of an opt.
Voltage ratio gives a fine first approximation.
Then you have "all the little details". We would like zero-Ohm windings but we only get windings with Ohms in them. We would like zero-loss iron but there is always some iron-loss, varying with frequency. At both ends of the audio band stray reactances start to get in the way.
The winding resistance is a series resistance. The iron loss is a shunt impedance. For a very rough estimate, both have similar effect. Say the nominal audio impedance is 10k. Say the winding resistance (reflected to primary) is 1k (1/10X). Now we see 11k. Say the iron loss is about 100k (10X). Now we see 9.9k which is ==10k so it was hardly worth figuring the losses.
The losses do matter in Power Output. If the tubes make 10 Watts, the above values lose about 10% in copper, about 10% in iron, so we only have 8 Watts to the speaker. That's 1dB loss, and that's actually fairly good for commercial OTs.
Lower loss is "possible" but getting lower winding resistance forces a much larger iron core to make space for fatter copper. The larger lump of iron has more iron-loss. The larger windings have more stray capacitance and this matters at plate impedance. These are not new questions (as shown by reference to RDH4). You can expect the good transformer makers to have done as good a job as possible for what they think the transformer should do.
If loss is low, figuring from Turns Ratio will always be very close to right. Remember that "same" tubes will be +/-20% in parameters so it is hardly worth using a too-sharp pencil. If driving loudspeakers, actual impedance is -20%/+500% (or worse), so we can only hope to be "close".
Fall-off of tube gain, power, distortion with impedance change is very gradual.
Voltage ratio gives a fine first approximation.
Then you have "all the little details". We would like zero-Ohm windings but we only get windings with Ohms in them. We would like zero-loss iron but there is always some iron-loss, varying with frequency. At both ends of the audio band stray reactances start to get in the way.
The winding resistance is a series resistance. The iron loss is a shunt impedance. For a very rough estimate, both have similar effect. Say the nominal audio impedance is 10k. Say the winding resistance (reflected to primary) is 1k (1/10X). Now we see 11k. Say the iron loss is about 100k (10X). Now we see 9.9k which is ==10k so it was hardly worth figuring the losses.
The losses do matter in Power Output. If the tubes make 10 Watts, the above values lose about 10% in copper, about 10% in iron, so we only have 8 Watts to the speaker. That's 1dB loss, and that's actually fairly good for commercial OTs.
Lower loss is "possible" but getting lower winding resistance forces a much larger iron core to make space for fatter copper. The larger lump of iron has more iron-loss. The larger windings have more stray capacitance and this matters at plate impedance. These are not new questions (as shown by reference to RDH4). You can expect the good transformer makers to have done as good a job as possible for what they think the transformer should do.
If loss is low, figuring from Turns Ratio will always be very close to right. Remember that "same" tubes will be +/-20% in parameters so it is hardly worth using a too-sharp pencil. If driving loudspeakers, actual impedance is -20%/+500% (or worse), so we can only hope to be "close".
Fall-off of tube gain, power, distortion with impedance change is very gradual.
Last edited:
it speaks to me, it's coherent,thank you PRR
I think that at one point, you have to stop making holes in your head and just enjoy doing things as simply as possible, but not just to mention a great person.
I understood what I wanted to understand, I will be able to continue to learn the rest and as it turns out that I accumulated several pairs of opt and a number of tubes, I have something to have fun during sometime .
I think that at one point, you have to stop making holes in your head and just enjoy doing things as simply as possible, but not just to mention a great person.
I understood what I wanted to understand, I will be able to continue to learn the rest and as it turns out that I accumulated several pairs of opt and a number of tubes, I have something to have fun during sometime .
For purists-perfectionists, please remember the impedance dependence on frequency curve for real loudspeakers, and say to yourself if 1-10% measurement error matters at all, especially if to add differences in tubes even manufactured in the same plant the same day.
- Status
- Not open for further replies.