mikelm said:
There are quite a few in the 50 to 60 volt range, plus some newer parts but would have to check in the morming for specific types.
BW,
Susan.
I just checked in spice.
The 20mH + 20mH choke is flat within 0.5db to 4hz @ 4A + 4A but the distortion starts to rise at about 100Hz
The 80mH + 80mH choke is flat within 0.5db to 1Hz @ 4A + 4A but the distortion stays flat until about 17Hz
I'll try both and see which I prefer
mike
The 20mH + 20mH choke is flat within 0.5db to 4hz @ 4A + 4A but the distortion starts to rise at about 100Hz
The 80mH + 80mH choke is flat within 0.5db to 1Hz @ 4A + 4A but the distortion stays flat until about 17Hz
I'll try both and see which I prefer
mike
Hi Mike,
Okay, interesting.
These are for distortion modeled at 10V peak into 8 ohms I assume.
=========
Example Automotive mosfets:
For Zeus amp running on a 12V system:
STP200NF04
N-CHANNEL 40V - 120A - 3.3mΩ - 310W - TO-220 STripFET™II MOSFET
http://www.st.com/stonline/books/pdf/docs/10073.pdf
For running on a 24V system:
STW220NF75
N-CHANNEL 75V - 120A - 4mΩ - 500W - TO-247 STripFET™II MOSFET
http://www.st.com/stonline/books/pdf/docs/9532.pdf
although here there are a lot more alternatives.
Spice models at:
http://www.st.com/stonline/prodpres/discrete/powmosft/spice/spice.htm
STP200NF04:
http://www.st.com/stonline/prodpres/discrete/powmosft/spice/txt/stp200nf04.txt
http://www.st.com/stonline/prodpres/discrete/powmosft/spice/lib/stp200nf04.lib
Have fun 🙂
Best wishes,
Susan.
mikelm said:
Okay, interesting.
These are for distortion modeled at 10V peak into 8 ohms I assume.
=========
Example Automotive mosfets:
For Zeus amp running on a 12V system:
STP200NF04
N-CHANNEL 40V - 120A - 3.3mΩ - 310W - TO-220 STripFET™II MOSFET
http://www.st.com/stonline/books/pdf/docs/10073.pdf
For running on a 24V system:
STW220NF75
N-CHANNEL 75V - 120A - 4mΩ - 500W - TO-247 STripFET™II MOSFET
http://www.st.com/stonline/books/pdf/docs/9532.pdf
although here there are a lot more alternatives.
Spice models at:
http://www.st.com/stonline/prodpres/discrete/powmosft/spice/spice.htm
STP200NF04:
http://www.st.com/stonline/prodpres/discrete/powmosft/spice/txt/stp200nf04.txt
http://www.st.com/stonline/prodpres/discrete/powmosft/spice/lib/stp200nf04.lib
Have fun 🙂
Best wishes,
Susan.
Site is up and running
http://www.susan-parker.co.uk is now back. This is a great site, the transformer design section is intersting. I was also looking at how Lundahl winds transformers, the five section winding in particular. http://www.lundahl.se/pdfs/papers/outp_typ.pdf
I have used the Lundahl transformers with excellent results in my 300B design. Would this winding style work well as opposed to windind wires 4 at a time in the same bundle.
The windings would look like this from a cross section view.
PPPPPPPPPPPPPP PPPPPPPPPPPPPP
---------------------- ----------------------
SSSSSSSSSSSSSS SSSSSSSSSSSSSS
---------------------- ----------------------
PPPPPPPPPPPPPP PPPPPPPPPPPPPP
---------------------- ----------------------
SSSSSSSSSSSSSS SSSSSSSSSSSSSS
---------------------- ----------------------
PPPPPPPPPPPPPP PPPPPPPPPPPPPP
http://www.susan-parker.co.uk is now back. This is a great site, the transformer design section is intersting. I was also looking at how Lundahl winds transformers, the five section winding in particular. http://www.lundahl.se/pdfs/papers/outp_typ.pdf
I have used the Lundahl transformers with excellent results in my 300B design. Would this winding style work well as opposed to windind wires 4 at a time in the same bundle.
The windings would look like this from a cross section view.
PPPPPPPPPPPPPP PPPPPPPPPPPPPP
---------------------- ----------------------
SSSSSSSSSSSSSS SSSSSSSSSSSSSS
---------------------- ----------------------
PPPPPPPPPPPPPP PPPPPPPPPPPPPP
---------------------- ----------------------
SSSSSSSSSSSSSS SSSSSSSSSSSSSS
---------------------- ----------------------
PPPPPPPPPPPPPP PPPPPPPPPPPPPP
Re: Site is up and running
Hi Mike,
Good, and thanks 🙂
The main reason for this complicated interleaving is the large difference in numbers of turns between the primaries and the secondaries. Plus the high voltages which have to be insulated from the output.
Yes, for my Zeus amplifiers one can wind them in this manner, but it doesn't give any advantages and the down side is the poorer performance and difficulty of construction.
Whilst sonically it might not seem to matter it is important to maintain as high as output stage bandwidth as possible as this give best follower action from the mosfets.
For a small number of turns with large diameter wire the multi-filar technique is better both in electrical terms and also from the point of actually making the beasts.
With a coil winding machine it is relatively easy to make these complex winding schemes, but it is still time consuming which reflects in the price one pays for the parts.
Making them by hand is well nigh impossible for a push-pull where one uses a double section bobbin and mirrored and matched windings (for a single ended design it isn't so difficult).
However with my designs using quad filar winding it is quite possible to make these output transformers by hand - literally holding the bobbin in one hand and winding the four wires with the other. All four windings are matched and close coupled.
Making the dual section bobbin is a little trickier, and it helps to have the bobbin held on a spindle with a turns counter, but still quite within the means of ordinary constructors (i.e. you and me).
I did make an output transformer with interleaved primaries and secondaries in the manner you describe for my sub-ohm speaker array project, but I still bifilar wound the primarys.
The input transformers on the other hand with their dual matched 1:10 sections are fiddly even with a three layer construction (1/2S😛:1/2S) per side and this is why I had mine made by Sowter. Unless I build myself a coil winding machine (one of many things planed but not yet got round to) I won't be making these myself.
I hope this helps.
Best wishes,
Susan
Hi Mike,
MIKET said:
Good, and thanks 🙂
The main reason for this complicated interleaving is the large difference in numbers of turns between the primaries and the secondaries. Plus the high voltages which have to be insulated from the output.
Yes, for my Zeus amplifiers one can wind them in this manner, but it doesn't give any advantages and the down side is the poorer performance and difficulty of construction.
Whilst sonically it might not seem to matter it is important to maintain as high as output stage bandwidth as possible as this give best follower action from the mosfets.
For a small number of turns with large diameter wire the multi-filar technique is better both in electrical terms and also from the point of actually making the beasts.
With a coil winding machine it is relatively easy to make these complex winding schemes, but it is still time consuming which reflects in the price one pays for the parts.
Making them by hand is well nigh impossible for a push-pull where one uses a double section bobbin and mirrored and matched windings (for a single ended design it isn't so difficult).
However with my designs using quad filar winding it is quite possible to make these output transformers by hand - literally holding the bobbin in one hand and winding the four wires with the other. All four windings are matched and close coupled.
Making the dual section bobbin is a little trickier, and it helps to have the bobbin held on a spindle with a turns counter, but still quite within the means of ordinary constructors (i.e. you and me).
I did make an output transformer with interleaved primaries and secondaries in the manner you describe for my sub-ohm speaker array project, but I still bifilar wound the primarys.
The input transformers on the other hand with their dual matched 1:10 sections are fiddly even with a three layer construction (1/2S😛:1/2S) per side and this is why I had mine made by Sowter. Unless I build myself a coil winding machine (one of many things planed but not yet got round to) I won't be making these myself.
I hope this helps.
Best wishes,
Susan
Winding machines
Hi Bill,
Thanks for your post 🙂
Yes I see what you mean.
I hadn't seen this one before. Looks handy.
I am using my lathe for the moment (unpowered) with a handle in the spindle, and a reflective opto triggering a timer/counter (which works well unless I forget and turn the handle backwards).
My setup isn't suitable for these hundreds of turns per section type transformer windings but does okay on the multi-filar ones.
I bought a big stepper to make the winding machine - it's a Pac Sci Sigmax Stepper motor K42HSFK-LEK-SS-02, NEMA 42 frame mounting, double stack, 4025 oz-in holding torque. 🙂
The bobbin holder will go straight on to the shaft, and a second smaller stepper with a short lead screw to control the wire position. A little bit of code on a PC and off we go! Well that's the theory anyway.
Will you let us know if you succumb to temptation?
Best wishes,
Susan.
Hi Bill,
Thanks for your post 🙂
Bill F. said:
Yes I see what you mean.
I hadn't seen this one before. Looks handy.
I am using my lathe for the moment (unpowered) with a handle in the spindle, and a reflective opto triggering a timer/counter (which works well unless I forget and turn the handle backwards).
My setup isn't suitable for these hundreds of turns per section type transformer windings but does okay on the multi-filar ones.
I bought a big stepper to make the winding machine - it's a Pac Sci Sigmax Stepper motor K42HSFK-LEK-SS-02, NEMA 42 frame mounting, double stack, 4025 oz-in holding torque. 🙂
The bobbin holder will go straight on to the shaft, and a second smaller stepper with a short lead screw to control the wire position. A little bit of code on a PC and off we go! Well that's the theory anyway.
Will you let us know if you succumb to temptation?
Best wishes,
Susan.
Certainly, but now I think I like your idea a lot better!
I'm setting up a large 3-axis CNC router table. (Well, that is to say I have all the pieces, but only the table/work surface is together so far). Maybe my software and controller could moonlight as the front end for a winding machine...
Will you let us know how your winder turns out? 😎
Hi Bill,
Oops, what have I started?
What size are you building and is it for wood or metal?
I had a similar plan using the controller I built for my Emco F1 milling machine (which I refurbished with new steppers as I just got the mill itself without any control bits).
3 off API CMD260s and a Minarik 180 volt DC regenerative drive for the spindle (not shown in pic - it goes in the drive bay), with a big 1200W SM-PSU.
One possibility is to program it with G-code as if one was setting up a lathe? Probably need a bit more sophistication but could get one going. I was thinking of merging right and left handed threading sequences and putting it in a loop for the number of turns.
Needs a bit more thought...
Proper control with a dedicated micro (a BlackFin STAMP board running uClinux and networked?) might be easier in the longer term.
I will let you know any progress - but please don't hold your breath as I am working on pre-amps at the moment🙂
Best wishes,
Susan.
Bill F. said:
Oops, what have I started?
What size are you building and is it for wood or metal?
I had a similar plan using the controller I built for my Emco F1 milling machine (which I refurbished with new steppers as I just got the mill itself without any control bits).
3 off API CMD260s and a Minarik 180 volt DC regenerative drive for the spindle (not shown in pic - it goes in the drive bay), with a big 1200W SM-PSU.
One possibility is to program it with G-code as if one was setting up a lathe? Probably need a bit more sophistication but could get one going. I was thinking of merging right and left handed threading sequences and putting it in a loop for the number of turns.
Needs a bit more thought...
Proper control with a dedicated micro (a BlackFin STAMP board running uClinux and networked?) might be easier in the longer term.
I will let you know any progress - but please don't hold your breath as I am working on pre-amps at the moment🙂
Best wishes,
Susan.
It'll have a work area about 3' x 5.5' x 6" With a footprint of 4' x 8' it takes up a horrendous chunk of shop space. It's a modest recipe of ballscrews, supported linear slides, 250oz/in. steppers, a simple unipolar drive, and an off-the-shelf wood-router spindle. It's for wood and maybe swivel-knifing vinyl.
Fun stuff. 🙂
I was thinking more along the lines of horns/waveguides/complex baffle contours cut from exotic hardwoods... but the sky's the limit. 🙂
(Sorry about the OT digression--we now return to the regularly scheduled programming.
)
(Sorry about the OT digression--we now return to the regularly scheduled programming.
)
Not wishing to shock too much but I will also be using wood .... for the entire enclosure - except audio hardwear.
I have always enjoyed working with wood a n d - wait for it - I think it helps with the sound to limit metal to a bare minimum... shock horror
cheers
mike
I have always enjoyed working with wood a n d - wait for it - I think it helps with the sound to limit metal to a bare minimum... shock horror
cheers
mike
Zeus Preamp - Line Driver
Dear All,
After much thought over the last couple of months and discusions with several people (thanks for your input 🙂 ) here is a sketch of a version of the Zeus pre-amp / line driver that uses the Stevens & Billington TX102 Magnetic Volume Control.
A PDF is here:
http://www.susan-parker.co.uk/zeus-pre-line-amp-sb-tx102-v1a.pdf
This is the pre-amp as shown at:
http://www.susan-parker.co.uk/zeus-line-driver-1.htm
with the Gardners 1:4 transformer replaced with the TX102s. I expect similar (or better) performance. Note as drawn it is still configured for 600 ohm use.
I am looking at options like using reed relays for the switching and input selection which isn't shown at the moment.
This design could also be used as a headphone amplifier.
And yes, I will be buying a pair to try out on the bench to sort out the details.
Meanwhile any comments, particually from those who are already using these parts, would be most welcome 🙂
Best wishes,
Susan.
Dear All,
After much thought over the last couple of months and discusions with several people (thanks for your input 🙂 ) here is a sketch of a version of the Zeus pre-amp / line driver that uses the Stevens & Billington TX102 Magnetic Volume Control.
A PDF is here:
http://www.susan-parker.co.uk/zeus-pre-line-amp-sb-tx102-v1a.pdf
This is the pre-amp as shown at:
http://www.susan-parker.co.uk/zeus-line-driver-1.htm
with the Gardners 1:4 transformer replaced with the TX102s. I expect similar (or better) performance. Note as drawn it is still configured for 600 ohm use.
I am looking at options like using reed relays for the switching and input selection which isn't shown at the moment.
This design could also be used as a headphone amplifier.
And yes, I will be buying a pair to try out on the bench to sort out the details.
Meanwhile any comments, particually from those who are already using these parts, would be most welcome 🙂
Best wishes,
Susan.
Re: Zeus Preamp - Line Driver
Well, well, well...
Guess what I have sitting here on my desk?
(sorry about the poor pics - camera phone)
These suckers are like hand grenades, much bigger than I expected. And heavy! This linestage is exactly what I need to drive 10+m balanced connects.
Susan-Parker said:
Well, well, well...
Guess what I have sitting here on my desk?
(sorry about the poor pics - camera phone)
These suckers are like hand grenades, much bigger than I expected. And heavy! This linestage is exactly what I need to drive 10+m balanced connects.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Re: Zeus Preamp - Line Driver
Hi Darkmobius,
I trust you are keeping well.
Well, this is indeed fortutious 🙂
Pics are fine, show all that is needed.
Look forward to hearing your impressions etc.
As I mentioned previously I am buying a pair myself but with lead times I don't expect to have them for at least a couple of weeks.
I better check with SB to see what the actual delivery schedules are going to be after that as using a pair per channel I will need another set once I move off the bench for proper listening tests.
They are as the ads say "reasuringly expensive" 🙂 , so I wanted to check this circuit out first to make sure I wasn't missing something.
Best wishes,
Susan.
Hi Darkmobius,
I trust you are keeping well.
darkmoebius said:
Well, this is indeed fortutious 🙂
Pics are fine, show all that is needed.
Look forward to hearing your impressions etc.
As I mentioned previously I am buying a pair myself but with lead times I don't expect to have them for at least a couple of weeks.
I better check with SB to see what the actual delivery schedules are going to be after that as using a pair per channel I will need another set once I move off the bench for proper listening tests.
They are as the ads say "reasuringly expensive" 🙂 , so I wanted to check this circuit out first to make sure I wasn't missing something.
Best wishes,
Susan.
Re: Re: Zeus Preamp - Line Driver
Uh, oh....
I missed that "little" detail when I briefly looked at the schematic. Well, let's just say that's a little out of the reach of my pocketbook right now. Two was bad enough(I'm buying a full turntable system right now).
I got lucky and bought them for less that 1/2 price unused from a guy here on DiyAudio who didn't have the time to build all his kits.
Vic
Susan-Parker said:
Uh, oh....
I missed that "little" detail when I briefly looked at the schematic. Well, let's just say that's a little out of the reach of my pocketbook right now. Two was bad enough(I'm buying a full turntable system right now).I got lucky and bought them for less that 1/2 price unused from a guy here on DiyAudio who didn't have the time to build all his kits.
Vic
Re: Zeus Preamp - Line Driver
Hi,
Sorry to dampen your enthusiasm.
If you had the oppertunity would you be able to wire up the pair as show in my schematic - but without the attenuation swiching - and hard-wire say -3 dB of something and see how they sound to you driving an output stage?
I agree that the price is in the "take a deep breath and exhale slowly" range. However one might otherwise need another gain stage and so it is perhaps not "quite" so expensive.
I guess also that the dollar exchange rate doesn't help.
For a lower cost alternative the TI headphone chip has proven to be good for getting things going and I have one of PA's QRV-07 PCBs to build up as a propper preamp.
http://home.swipnet.se/~w-50719/hifi/qrv07/startsida_qrv07.html
I also have a QRV-06 diamond buffer PCB to build and try out for comparison (I have good hopes of a nice result with this).
http://home.swipnet.se/~w-50719/hifi/qrv06/startsida_qrv06.html
The great thing with building ourselves is that we have these choices 🙂
Best wishes,
Susan.
Hi,
darkmoebius said:
Uh, oh....
I got lucky and bought them for less that 1/2 price unused from a guy here on DiyAudio who didn't have the time to build all his kits.
Vic
Sorry to dampen your enthusiasm.
If you had the oppertunity would you be able to wire up the pair as show in my schematic - but without the attenuation swiching - and hard-wire say -3 dB of something and see how they sound to you driving an output stage?
I agree that the price is in the "take a deep breath and exhale slowly" range. However one might otherwise need another gain stage and so it is perhaps not "quite" so expensive.
I guess also that the dollar exchange rate doesn't help.
For a lower cost alternative the TI headphone chip has proven to be good for getting things going and I have one of PA's QRV-07 PCBs to build up as a propper preamp.
http://home.swipnet.se/~w-50719/hifi/qrv07/startsida_qrv07.html
I also have a QRV-06 diamond buffer PCB to build and try out for comparison (I have good hopes of a nice result with this).
http://home.swipnet.se/~w-50719/hifi/qrv06/startsida_qrv06.html
The great thing with building ourselves is that we have these choices 🙂
Best wishes,
Susan.
Preamp - Line Driver transformers
Dear All,
I have received details and pricing from Sowter for the Preamp - Line Driver output transformers.
Based on a valve interstage type it is 4 x 150 turns quadfilar on a 2in/#29 former and to the following dimensions.
Price is UKP 69.29 ( UKP 84.24 with VAT ) each.
Although this size transformer may seem overkill the mosfets are biased for at least 250 mA per device so there is some current flowing even when "at rest" and the winding wire used must therefor be of sufficient gauge for this (in this case it is 0.63 mm which has a max recommended current of 900 mA).
Mosfets should be matched as well as possible but pragmatically it is not possible always to get this absolutely perfect so there may be a mA or two still flowing and the larger core size will prevent saturation problems.
And this output stage will drive headphones down to 32 ohms.
Suitable input transformers for the striaght line stage as shown above, i.e. without a steped TVC, from Sowter are:
10K input impedance Input Transformer:
============================
Sowter Type 8920, 10k:10k+10k Input Phase Splitter Transformer 100% balance, UKP 44.76 plus VAT each (from Sowter price list).
http://www.sowter.co.uk/specs/8920.htm
This gives a 1:2 step up but should be drivable by any phono output.
600 ohm input impedance Input Transformer:
============================
Sowter Type 9063, 600:10k Very high level input. 1CT:2+2 UKP 60.04 plus VAT each (from Sowter price list).
http://www.sowter.co.uk/specs/9063.htm
This gives a 1:4 step up but needs to be driven from a proper 600 ohm line output.
Hope this helps.
Best wishes,
Susan.
Dear All,
I have received details and pricing from Sowter for the Preamp - Line Driver output transformers.
Based on a valve interstage type it is 4 x 150 turns quadfilar on a 2in/#29 former and to the following dimensions.
Price is UKP 69.29 ( UKP 84.24 with VAT ) each.
Although this size transformer may seem overkill the mosfets are biased for at least 250 mA per device so there is some current flowing even when "at rest" and the winding wire used must therefor be of sufficient gauge for this (in this case it is 0.63 mm which has a max recommended current of 900 mA).
Mosfets should be matched as well as possible but pragmatically it is not possible always to get this absolutely perfect so there may be a mA or two still flowing and the larger core size will prevent saturation problems.
And this output stage will drive headphones down to 32 ohms.
Suitable input transformers for the striaght line stage as shown above, i.e. without a steped TVC, from Sowter are:
10K input impedance Input Transformer:
============================
Sowter Type 8920, 10k:10k+10k Input Phase Splitter Transformer 100% balance, UKP 44.76 plus VAT each (from Sowter price list).
http://www.sowter.co.uk/specs/8920.htm
This gives a 1:2 step up but should be drivable by any phono output.
600 ohm input impedance Input Transformer:
============================
Sowter Type 9063, 600:10k Very high level input. 1CT:2+2 UKP 60.04 plus VAT each (from Sowter price list).
http://www.sowter.co.uk/specs/9063.htm
This gives a 1:4 step up but needs to be driven from a proper 600 ohm line output.
Hope this helps.
Best wishes,
Susan.
