Because in higher RF frequencies does exist two resistances, one "AC resistance" caused by skin and proximity effects in the wire, and a "DC resistance" measured at DC or very low frequencies. Both adds and contribute to the total copper losses in an inductor or transformer.
Search for "slup125" in the google engine, from Lloyd Dixon at TI.
Search for "slup125" in the google engine, from Lloyd Dixon at TI.
That is incredible.
We consider that the difference in a Rdc windings, as reported from me some post ago, is about 10 ohm so due to this value the distortion increase???
I post some post ago the picture in attach, it is a equivalent circuit at mid frequency of a o.t. trafo from RDH, just to have a simple example . (nobody sent a comment on it).
We can see it is a partition with two resistors where Rw is the Rdc of windings and RL is the secondary load resistence reffered to primary, this means that if I have 8 ohm on secondary on primary I will have 3200 ohm because the ratio is 20:1.
The Rw is 12 ohm , the Rdc difference between the two trafo ( normal and Litz, normal has 80 ohms and Litz has 92 ohm) so we cosider this one in our calculation of losses to get the differences between wires.
This is less than 0,1 dB!!!!
Do you think it is dangerous!!! Do you think the THD increase??
Increasing the frequencies other problems comes up, of course, but it is sure that with Litz the R (dc+skin) will increase lower than normal wire!!!!!
Walter
We consider that the difference in a Rdc windings, as reported from me some post ago, is about 10 ohm so due to this value the distortion increase???
I post some post ago the picture in attach, it is a equivalent circuit at mid frequency of a o.t. trafo from RDH, just to have a simple example . (nobody sent a comment on it).
We can see it is a partition with two resistors where Rw is the Rdc of windings and RL is the secondary load resistence reffered to primary, this means that if I have 8 ohm on secondary on primary I will have 3200 ohm because the ratio is 20:1.
The Rw is 12 ohm , the Rdc difference between the two trafo ( normal and Litz, normal has 80 ohms and Litz has 92 ohm) so we cosider this one in our calculation of losses to get the differences between wires.
This is less than 0,1 dB!!!!
Do you think it is dangerous!!! Do you think the THD increase??
Increasing the frequencies other problems comes up, of course, but it is sure that with Litz the R (dc+skin) will increase lower than normal wire!!!!!
Walter
Last edited:
I sincerely don't know if it measurable the distortion created by the frequency dependant resistance of the winding, but in any case I believe that is of negligible importance. In such a case, it adds to the needless of litz wire to wind an audio transformer excepting the case you have several rolls of it wasting space at your warehouse.
You confirmed more than once that you have used the same bobbins for the normal wire transformer and the litz wire transformer.
I don't believe it!
The EI normal wire transformer seems quite a bit smaller than the double c-core litz wire transformer; the bobbins cannot be the same.
When you would have used the same bobbins for both transformers and equal number of turns with equal winding geometry (which I also don't believe) the difference will be larger than 12 ohms.
Which size has the EI transformer? Something like EI 120?
Which c-cores are used? Something in the SG or SE series?
Your numbers for the EI transformer of post#1 are wrong, physically impossible unless you use a superconducting copper.
Ok,
my error, are 40 ohms more!!!
So we can't reach again the 0,1 dB of loss
ciao
Walter
Calculations show about 120Ω for the EI transformer, 40Ω more...
Please remove the covers and take another picture with both transformers so the mystery will solved.
Last edited:
Thank for the reference, but if I'm not mistaken it also discusses the benefit of Litz in reducing eddy current losses?
Too busy at work to chase down if it's meaningful here..
THD for silicon steel has been calculated by Dr.N.Partridge (see RDH4, p.215)
Vh/Vf = [(Shx10⁹ lm Ra)/(8 π² Np² S f)] [1 - (Ra/4 Zf)]
Where
Vh = Harmonic voltage across the primary
Vf = Fundamental voltage across the primary
Sh = Distortion coefficient of the magnetic material
lm = Lenght of the magnetic path
Np = Number of primary turns
Ra = Resistance (or equivalent resistance) in series with the primary
S = Cross-sectional area of the core
f = Fundamental frequency in Hz
Zf = Primary impedance at fundamental frequency ≈ 2 π f L
L = Primary inductance at chosen field B
Clearly THD increases with Ra, even in simulations it can be seen.
Walter,
From your first post #1 you indicated the litz you were using was 10 strand, 0.3mm2, which is equivalent to 0.62mmD solid.
But you didn't indicate whether that was just with enamelled strands, or enamelled with extra insulation such as silk, or with outer insulation, and whether there was any bundling of strands, or significant pitch, or the nominal OD of the wire. Do you have a link to the wire spec?
Did you look at other options for litz, as there are many stock items available for the switchmode market that are somewhat simple bundles with just enamelled strands, all the way to very thin outer teflon insulation sheath for standards rated insulation.
As I see it, various winding wire types - all with the same OD - could be compared in the same transformer (same core, bobbin, layers etc). The differences would manifest themselves as measureable DC resistance, AC resistance, leakage inductance, self capacitance, and transfer Q shape. If dominant distortion was a function of plate resistance, then the effective winding resistance would influence the outcome.
Modern tech improvements have no doubt helped transformer makers push their high frequency performance through materials such as Nomex, and higher temperature materials for baking out moisture prior to sealing the winding. Did you look at teflon coated winding wire?
Normally exotic materials are shunned, or deemed too impractical for audio applications, but you seem to have a keeness for pursuing such things.
From your first post #1 you indicated the litz you were using was 10 strand, 0.3mm2, which is equivalent to 0.62mmD solid.
But you didn't indicate whether that was just with enamelled strands, or enamelled with extra insulation such as silk, or with outer insulation, and whether there was any bundling of strands, or significant pitch, or the nominal OD of the wire. Do you have a link to the wire spec?
Did you look at other options for litz, as there are many stock items available for the switchmode market that are somewhat simple bundles with just enamelled strands, all the way to very thin outer teflon insulation sheath for standards rated insulation.
As I see it, various winding wire types - all with the same OD - could be compared in the same transformer (same core, bobbin, layers etc). The differences would manifest themselves as measureable DC resistance, AC resistance, leakage inductance, self capacitance, and transfer Q shape. If dominant distortion was a function of plate resistance, then the effective winding resistance would influence the outcome.
Modern tech improvements have no doubt helped transformer makers push their high frequency performance through materials such as Nomex, and higher temperature materials for baking out moisture prior to sealing the winding. Did you look at teflon coated winding wire?
Normally exotic materials are shunned, or deemed too impractical for audio applications, but you seem to have a keeness for pursuing such things.
Last edited:
Distortion coefficient in old 195x - 196x valid only only for material manufactured at that time. For modern GOSS steel it is much much lower. In fact, it is so low, that can be discarded unless you drive core close to saturation.
Walter specified the primary DCR of the litz transformer being 92 ohm (post 126).
With 10 strand litz wire equivalent to 0.3mm² (and 0.62 mm diameter) that would mean that the length of the primary winding is more than 1600 meters.
If he would have used a massive core with a mean winding length of 40 cm (which does not seem to be the case) he would have 4000 primary turns.
The 10 strand litz wire will have an OD of not less than 0.8-0.9 mm, so it would never be possible to get that amount of turns on a big core together with the secondary (needing about the same winding space as the primary) and the isolation layers.
His specifications are not reliable to say the least; therefore this discussion is meaningless
There would be some 'pitch' factor for litz DCR, indicating primary length could be noticeably less than a solid wire 'length' estimate. And there is 7 strand litz with basic insulation rated outer sheath and only 0.83mm OD. Perhaps let's wait for Walter before turning our noses up at the discussion.
- Status
- Not open for further replies.