I have been thinking recently about attempting a Tango 300EQ style riaa. At this point, I’m thinking a 6S45PE front end, transformer coupled to the LCR eq.
This brings me to my question…what are the potential pitfalls of winding a interstage xfrmer compared to a output xfrmer ? I’m considering a core of m6 with .3mm laminations with a dc saturation of around .5T.
Any thoughts ?
This brings me to my question…what are the potential pitfalls of winding a interstage xfrmer compared to a output xfrmer ? I’m considering a core of m6 with .3mm laminations with a dc saturation of around .5T.
Any thoughts ?
I've not yet tried !
Appart from using a thinner and most fragile wire, I beleive the main difference is that, in an OPT, both sides of the secondary are seen as "ground" from the primary and this could dictate a different interleaving scheme.
Perhaps you already know that paper from TI:
http://www-s.ti.com/sc/techzip/slup222.zip
It's related to SMPS transfos but has clear, interrsting and sometimes unnatempted analisys about leakage inductance (see part 3).
Yves.
Appart from using a thinner and most fragile wire, I beleive the main difference is that, in an OPT, both sides of the secondary are seen as "ground" from the primary and this could dictate a different interleaving scheme.
Perhaps you already know that paper from TI:
http://www-s.ti.com/sc/techzip/slup222.zip
It's related to SMPS transfos but has clear, interrsting and sometimes unnatempted analisys about leakage inductance (see part 3).
Yves.
Thanx for the reply Yves,
I’m pretty sure there is more to it, but, wouldn’t it basically be a low power SE OPT with a higher than usual secondary impedance. For example, if I use the 5K loadline on the 6S45PE,like the el-spuddo amp for example, and I plan on a 600ohm LCR, wouldn’t I just increase the secondary impedance. I realize there is a heck of a lot more to it for a true low level xfrmer ( as much magic as science from what I can tell) , but I would be dealing with an approx. 200mV rms signal (5mV X mu 40) on the primary. And since the core is partially saturated from the SE operation of the tube, I believe I would be operating in the linear portion of the hysterisis loop of the core.
If anybody could point out where I’m missing something PLEASE let me know (I have that nagging feeling I’m overlooking something obvious).
Thanx,
Casey
I’m pretty sure there is more to it, but, wouldn’t it basically be a low power SE OPT with a higher than usual secondary impedance. For example, if I use the 5K loadline on the 6S45PE,like the el-spuddo amp for example, and I plan on a 600ohm LCR, wouldn’t I just increase the secondary impedance. I realize there is a heck of a lot more to it for a true low level xfrmer ( as much magic as science from what I can tell) , but I would be dealing with an approx. 200mV rms signal (5mV X mu 40) on the primary. And since the core is partially saturated from the SE operation of the tube, I believe I would be operating in the linear portion of the hysterisis loop of the core.
If anybody could point out where I’m missing something PLEASE let me know (I have that nagging feeling I’m overlooking something obvious).
Thanx,
Casey
Interstage coupling transformers are just a bit different beast...
SInce you "usually" are not reflecting a resistance from secondary to primary.... It depends...
First step is to just consider the primary side as a choke...and get your specs lined up before the design begins...
At mid-band the gain of the stage is mu of the tube......
You basically calculate what INDUCTANCE you need based on the low frequency response you would like.... this based on plate resistance..
Then you calculate the maximum acceptable winding capacitance and leakage for the given high frequency response you want....also depends on plate resistance...
Now to look at the step response of the circuit.... You have the plate resistance acting as a damping R to this tank circuit....
If this TRIODE is low enough plate resistance to give a good response without overshoot, ect...fine... If it is not, then you would have to load the secondary of the transformer at the grid of the next stage and when that reflects into the primary it must satisfy the damping response you intend...
ALso....you need to select what turns ratio is appropriate for the over-all voltage gain of this stage...
Setting up the GAP and fux density equations is next.....
You have two flux density equations to work with...The DC flux density which is based on the GAP, DC current and number of turns... Then you have the AC flux density which is based on the AC rms signal that will across the winding, AREA, TURNS, lowest frequency of interest 20Hz....
WHen you go Single Ended with M6 your AC flux + DC flux should max out about 13,000 Gauss to 14,000 Gauss... SO for starters you could split the Flux evenly between the AC and DC flux and do 6500 Gauss for each.... You will have to re-iterate this process till you converge upon satisfying your specs as best as possible..
Experience usually makes this a quicker process...
Chris
SInce you "usually" are not reflecting a resistance from secondary to primary.... It depends...
First step is to just consider the primary side as a choke...and get your specs lined up before the design begins...
At mid-band the gain of the stage is mu of the tube......
You basically calculate what INDUCTANCE you need based on the low frequency response you would like.... this based on plate resistance..
Then you calculate the maximum acceptable winding capacitance and leakage for the given high frequency response you want....also depends on plate resistance...
Now to look at the step response of the circuit.... You have the plate resistance acting as a damping R to this tank circuit....
If this TRIODE is low enough plate resistance to give a good response without overshoot, ect...fine... If it is not, then you would have to load the secondary of the transformer at the grid of the next stage and when that reflects into the primary it must satisfy the damping response you intend...
ALso....you need to select what turns ratio is appropriate for the over-all voltage gain of this stage...
Setting up the GAP and fux density equations is next.....
You have two flux density equations to work with...The DC flux density which is based on the GAP, DC current and number of turns... Then you have the AC flux density which is based on the AC rms signal that will across the winding, AREA, TURNS, lowest frequency of interest 20Hz....
WHen you go Single Ended with M6 your AC flux + DC flux should max out about 13,000 Gauss to 14,000 Gauss... SO for starters you could split the Flux evenly between the AC and DC flux and do 6500 Gauss for each.... You will have to re-iterate this process till you converge upon satisfying your specs as best as possible..
Experience usually makes this a quicker process...
Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.