So I have been planning to build this microphone for a while.. It is the GIC G7 microphone from : http://www.gyraf.dk/gy_pd/g7/gic.htm. They use the transformer as described "the wrong way arround" to transform the 15K impedance of the EF86 to something more useable.
I dont see much mike talk here but my problem is in its roots general.
The transformer is simply too expensive or unobtainable from my country. I could buy something criminally overpriced from ebay, or go with a EDCOR transformer but their shipping is at least 3X the value of the part to my location. So I went with the decision "what the hell just try and wind one".
Unfortunately I have not yet gotten into building the thing but I have done some math on the transformer. The highest permeability core I can get locally would be 2700 (or AL=8700nH). So what I have worked out with a online calculator is that 135 turns would give me roughly 2H on a 13mm core with 0.1mm wire. The core would be RM tipe.
This would be the secondary which would be 600. Since I need a transformer with a ratio of 25 I now have to apply this to the turns ratio which is 3375(LOT OF TURNS BY HAND). With this many turns the primary tube side inductance would be 1200H.
That does not quite add up in my mind I just want to verify if that sounds right to other forum members and if possibly for any advice correct me on the matter of designing this transformer.
I did not quite like the result of my calculation so I wen to other manufacturers who have transformers (as an example TOROIDY with their toroidal output transformers) and as an example my EL34PP transformers I use are about 590H primary so with a little common sense it sounds just about believeable that my transformer would come out to be 1200H.
I could play arround with the wire thickness and eventually find the point where I use all the space of the bobbin but I dont think it would matter as long as the inductance of the primary side is the same the secondary will stay also the same. And I didnt account for layers of the wire aswell but with .1mm it shouldnt really matter.
I dont know really I am not sure I wonder whats the opinion of some more people onto this.
Technical data that I want to achieve:
Pri Z:15Kohm
Sec Z: 600ohm
Just to add the frequency bandwidth here I would expect this to perform in
30Hz up to 30Khz (the high end isnt all that critical as long as its above 22Khz)
I did something wrong here dont burn my house down please I am just a student in electronics and a wanderer in tube stuff. I can only make PCBs well.
For the people assuming: All I want is your opinion if it sound believeable. If you have any idea or advice to try I will gladly consider it and try it.
I dont see much mike talk here but my problem is in its roots general.
The transformer is simply too expensive or unobtainable from my country. I could buy something criminally overpriced from ebay, or go with a EDCOR transformer but their shipping is at least 3X the value of the part to my location. So I went with the decision "what the hell just try and wind one".
Unfortunately I have not yet gotten into building the thing but I have done some math on the transformer. The highest permeability core I can get locally would be 2700 (or AL=8700nH). So what I have worked out with a online calculator is that 135 turns would give me roughly 2H on a 13mm core with 0.1mm wire. The core would be RM tipe.
This would be the secondary which would be 600. Since I need a transformer with a ratio of 25 I now have to apply this to the turns ratio which is 3375(LOT OF TURNS BY HAND). With this many turns the primary tube side inductance would be 1200H.
That does not quite add up in my mind I just want to verify if that sounds right to other forum members and if possibly for any advice correct me on the matter of designing this transformer.
I did not quite like the result of my calculation so I wen to other manufacturers who have transformers (as an example TOROIDY with their toroidal output transformers) and as an example my EL34PP transformers I use are about 590H primary so with a little common sense it sounds just about believeable that my transformer would come out to be 1200H.
I could play arround with the wire thickness and eventually find the point where I use all the space of the bobbin but I dont think it would matter as long as the inductance of the primary side is the same the secondary will stay also the same. And I didnt account for layers of the wire aswell but with .1mm it shouldnt really matter.
I dont know really I am not sure I wonder whats the opinion of some more people onto this.
Technical data that I want to achieve:
Pri Z:15Kohm
Sec Z: 600ohm
Just to add the frequency bandwidth here I would expect this to perform in
30Hz up to 30Khz (the high end isnt all that critical as long as its above 22Khz)
I did something wrong here dont burn my house down please I am just a student in electronics and a wanderer in tube stuff. I can only make PCBs well.
For the people assuming: All I want is your opinion if it sound believeable. If you have any idea or advice to try I will gladly consider it and try it.
Last edited:
You may need to get one made with a range like that.
J30A11C | Through Hole Audio Transformer 3.75 kΩ, 15 kΩ | RS Components
is reasonable.
Your ears cannot hear 22kHZ; most people lose hearing above 17kHZ.
That would save some money having one made. Or use a buffer and go 600Ω - 600Ω with an ECC82 or better still 12AY7.
J30A11C | Through Hole Audio Transformer 3.75 kΩ, 15 kΩ | RS Components
is reasonable.
Your ears cannot hear 22kHZ; most people lose hearing above 17kHZ.
That would save some money having one made. Or use a buffer and go 600Ω - 600Ω with an ECC82 or better still 12AY7.
Well I am a young fella so its not really 17K and done 🙂. Besides id like to use this for more than just a demonstration.
The transformer you linked isnt the right ratio and to te budget I have VERY expensive... and the shiping again RS components no go and asking someone to make me one is again a no go since I dont expect anyone to do it cheaper than that.
The transformer you linked isnt the right ratio and to te budget I have VERY expensive... and the shiping again RS components no go and asking someone to make me one is again a no go since I dont expect anyone to do it cheaper than that.
Maybe the NTE10/3? That one is pretty inexpensive, and will probably work OK for the price..
/Jakob E.
/Jakob E.
Actually I am sorry. I didnt read the specs correctly and did not realise I could wire the windings up to get me the correct ratio. But its still very expensive. At least now I know that my estimation of 2H on the 600ohm side was good.
And if I were to go based on that I would end up really with the rediculous 1000H. I have done some check arround on mouser for similar stuff I want and probably that inductance will be okay.. I will buy the parts and wind something up on the core then see how bad it will turn out to be. I might consider going with thicker wire to possibly alter the inductance in my favour.
Thanks wow that looks a whole loet more affordable but I cant really find a datasheet for this.
So far my question wasnt asnwered at all. Does anyone have any idea if the primary 15K ohm side inductance is okay to be 1200H? Thats my only concern right now.
Thanks, Adam.
So I found my problem.
I didnt realise that impedance is squared the winding ratio. That means my inductance shouldnt be 25 times more than the secondary but 5 times. The winding turn ratio will be 1:5 and that makes everything easier.
In that case the primary turns out to be 650 turns. The estimate for the inductance would be 46H which is much much much more reallistic. Now I am confident to go ahead and do it. Ill see what results this will get me but I have high hopes for it despite nobody trying this before on a ferrite core (as far I know).
I didnt realise that impedance is squared the winding ratio. That means my inductance shouldnt be 25 times more than the secondary but 5 times. The winding turn ratio will be 1:5 and that makes everything easier.
In that case the primary turns out to be 650 turns. The estimate for the inductance would be 46H which is much much much more reallistic. Now I am confident to go ahead and do it. Ill see what results this will get me but I have high hopes for it despite nobody trying this before on a ferrite core (as far I know).