NP Acoustics is indeed still going. The winder isn't that old.
Nickel seems the way to go here - can you guys recommend any OPTs that are part nickel?
@Grid01 where in Japan do you buy the nickel laminations from? I don't need a custom winder.
IMHO, the core type is the with 45%-to-50% nickel. There is a small number of them with slightly different amount of Nickel (and are called with different names like Radiometal 4550, High Perm 49/4750) and 50% iron. They still have very low loss and high permeability relative to SiFe GOSS but they have saturation around 1.5T!
Respect to 78%-80% Nickel alloys basically they trade a small amount of loss and some permeability to achieve higher saturation the latter being pretty poor in high nickel content alloys for use in output transformers. It's acceptable until the power level is low (and the OPT is big in comparison) but not practical if the power level grows, especially if the power rating is truly specified at 30Hz or less.
The other advantage of Ni-Fe alloys is that they are not oriented and so they don't have the disadvantage of EI GOSS laminations that are only 2/3 oriented (which means higher loss). This gives a lot of flexibility on size vs core section. One can buy 1 or 2 sizes for many applications. That's not possible with C-cores. The square shape of EI cores respect to C cores has a small disadvantage at high frequency but 100KHz bandwidth is no problem anyway....
IMHO, the core type is the with 45%-to-50% nickel. There is a small number of them with slightly different amount of Nickel (and are called with different names like Radiometal 4550, High Perm 49/4750) and 50% iron. They still have very low loss and high permeability relative to SiFe GOSS but they have saturation around 1.5T!
Respect to 78%-80% Nickel alloys basically they trade a small amount of loss and some permeability to achieve higher saturation the latter being pretty poor in high nickel content alloys for use in output transformers. It's acceptable until the power level is low (and the OPT is big in comparison) but not practical if the power level grows, especially if the power rating is truly specified at 30Hz or less.
The other advantage of Ni-Fe alloys is that they are not oriented and so they don't have the disadvantage of EI GOSS laminations that are only 2/3 oriented (which means higher loss). This gives a lot of flexibility on size vs core section. One can buy 1 or 2 sizes for many applications. That's not possible with C-cores. The square shape of EI cores respect to C cores has a small disadvantage at high frequency but 100KHz bandwidth is no problem anyway....
One can get around that using a larger core, until the power required is too high but certainly possible up to 20-25W @30Hz. Bigger size and likely more weight....
The reason why I don't like Metglas is that such material is brittle, lots of cores I have seen for sale look awful and they still ask for a lot of money. It also explains why the MM transformers with that core are outrageously expensive, they have to make/require a lot of core selection.
One can get the same specs by using Ni-Fe alloys, just tuning the amount of Ni. Differences in properties will be so small that they will eventually vanish in SE applications where the airgap dominates.
Last edited:
If you want a supply from Japan you will have to pay for the stamp for the size and more suitable for smaller parafeed transformers again expensive 5K plus .
For the Ei96 lams I got from South Korea until he had sold out of his supply and now you would have to go to the factory 100kg
<Quotation>
1. P/N : PB 96EI 0.35T 6mm holes
2. Q’ty : 100Kg
3. Price : USD70.00/Kg based on the FOB Korean Sea Port
4. Payment terms : T/T in advance before shipment
5. Lead time : 3weeks after getting P.O
For a one of pair you would be better to go to Dave Slagle
For the Ei96 lams I got from South Korea until he had sold out of his supply and now you would have to go to the factory 100kg
<Quotation>
1. P/N : PB 96EI 0.35T 6mm holes
2. Q’ty : 100Kg
3. Price : USD70.00/Kg based on the FOB Korean Sea Port
4. Payment terms : T/T in advance before shipment
5. Lead time : 3weeks after getting P.O
For a one of pair you would be better to go to Dave Slagle
Last edited:
I will make a plug for parallel feed output transformers again. When you keep the DC out of the transformer, you don't need the air gap and the transformer can be a lot smaller, keeping the material cost down for more expensive metals, such as nickel. Also the parallel feed setup allows for wiring the transformer as an autoformer, which I believe to be more transparent than conventional setup's.
In my case, my parallel feed opt is permalloy for a 300b output type handling above 500 Hz, so I use a more transparent V-cap for the parallel feed cap. What's not to like?
In my case, my parallel feed opt is permalloy for a 300b output type handling above 500 Hz, so I use a more transparent V-cap for the parallel feed cap. What's not to like?
Well, the extra cap in the path of the signal (and added resonance which can be dealt with but needs to be carefully taken into account) + the transformer is not more transparent than the Nickel transformer only and the extra cost of a quality signal cap does not justify it. That's my opinion.
The 45-50% nickel core will be basically equal to what I normally use with EI GOSS lamination because EI GOSS lamination cannot work at high B like C cores, despite on paper those of best quality saturate at 1.7T (formally). For best quality I stop at 0.85T full power @30Hz. That is possible with 45-50% nickel too.
Mumetal is good for signal xformers where is unbeatable, IMHO.
Retsel,
I am curious about your 300B Parafeed amplifier.
Would you Post a complete and accurate schematic of your 300B Parafeed amplifier, please?
Thanks!
Are you using a Choke as the CCS Plate Load for the 300B?
Or . . .
Are you using an electronic CCS Plate Load for the 300B?
Such as a Vacuum Tube CCS, IC CCS, BJT CCS, or MOSFET CCS for the 300B Plate load?
This requires up to 2x the B+, Versus a Choke CCS Plate Load.
How much Signal loss does a CCS Choke Plate Load have?
(versus the Signal loss of a near perfect Electronic CCS)?
Not the obvious loss at low frequencies due to the Choke inductance.
I am talking about the mid frequency and high frequency Signal loss due to the Choke"s laminations, and the Choke"s distributed capacitance.
The only times I built Parafeed outputs, I used a Choke CCS plate load.
Please, can any body who has tried Both a Choke plate load, And an Electronic plate load, tell us the difference in sound and the difference in measurements between the two types of CCS plate load; Choke versus Electronic?
A Challenge: Food For Thought:
In order to not have to use 2x B+ Voltage . . .
Could certain materials for laminations be used to an advantage, for a Parafeed CCS Choke Plate Load?
(Versus using SiFe laminations)?
Thanks!
If special laminations is the way to build a Parafeed Choke Plate Load, I might be willing to try Parafeed again.
I am curious about your 300B Parafeed amplifier.
Would you Post a complete and accurate schematic of your 300B Parafeed amplifier, please?
Thanks!
Are you using a Choke as the CCS Plate Load for the 300B?
Or . . .
Are you using an electronic CCS Plate Load for the 300B?
Such as a Vacuum Tube CCS, IC CCS, BJT CCS, or MOSFET CCS for the 300B Plate load?
This requires up to 2x the B+, Versus a Choke CCS Plate Load.
How much Signal loss does a CCS Choke Plate Load have?
(versus the Signal loss of a near perfect Electronic CCS)?
Not the obvious loss at low frequencies due to the Choke inductance.
I am talking about the mid frequency and high frequency Signal loss due to the Choke"s laminations, and the Choke"s distributed capacitance.
The only times I built Parafeed outputs, I used a Choke CCS plate load.
Please, can any body who has tried Both a Choke plate load, And an Electronic plate load, tell us the difference in sound and the difference in measurements between the two types of CCS plate load; Choke versus Electronic?
A Challenge: Food For Thought:
In order to not have to use 2x B+ Voltage . . .
Could certain materials for laminations be used to an advantage, for a Parafeed CCS Choke Plate Load?
(Versus using SiFe laminations)?
Thanks!
If special laminations is the way to build a Parafeed Choke Plate Load, I might be willing to try Parafeed again.
Last edited:
You can pick up any supplier you like from China on Alibaba. They ship worldwide. BTW, I would suggest if you order amorphous or nano-crystalline add an option - epoxy coating.You must tell me where you got them and the cost!
Attachments
Thanks. Do you find the quality to be that good or do you need to specify it? I am especially thinking about the cut and the matching.
Why not just make the cores clamped between the two metal plates wider to fill in the gap in the coil former? How is the pictured design better than that?
There are just a handful of amorphous/nano ribbon manufacturers in China, so you don't have much choice anyway no matter how many core suppliers you find. Main were from from Anhui Lizhi Magnetic Material. Once you air-gap them, any discrepancy in permeability will be insignificant. I tested amorphous and nano in push-pull output transformer, and sold out all pairs of amorphous, left only 1 pc of unpaired nano. Decided to stick with HiB cores, finished product quality is roughly same, yet GOSS HiB are cheaper and smaller for same level of flux density.
Mine are certainly not the best of all for 300B SE, but they do the job fine, and allowed me to build compact 300B mono-block :
Model HH-25-B
ZP = 2000R/2500R.
ZS = 4R/8R/16R.
Nom. Pwr = 25WRMS.
I prim max = 100mA
C-type circuit.
weight = 1.5kg
These were made in France by STS Millerioux, but this company, after been bought out, has closed its doors since.
T
Model HH-25-B
ZP = 2000R/2500R.
ZS = 4R/8R/16R.
Nom. Pwr = 25WRMS.
I prim max = 100mA
C-type circuit.
weight = 1.5kg
These were made in France by STS Millerioux, but this company, after been bought out, has closed its doors since.
T