That's an excellent unit LoudThud. Why is it that Yanks get all of this stuff on their doorstep and we seem to languish? I had no problems with how the autotransformer works and, as my output is already isolated, it has always seemed to be the best solution. The only issue was details of practical implementation. That said, I came across this site:
Turner Audio Transformer Winding
This guy knows his onions! Trouble is it takes some time to get through the stuff he's talking about but it's a Godsend and I'll be digesting it soon.
For this application, if I'm going to wind one I'll use a good quality core suited to audio frequency range work Andrew. For anyone interested in this sort of stuff, I would recommend having a look at this range:
Silicon Steel Toroidal O-Cores
O-core is the Holy Grail of this sort of stuff. Don't forget that this is wide range GOSS material in the best shape around. They offer a range of sizes and all incredibly cheap. The OA80 which is a 96-182VA unit, comes out at $14 but the postage to the UK puts the price up $123, not viable no matter how good they are!
I emailed Alpha-Core about the postage cost and got a reply back with an offer to ship their GOSS Toroidal O-core units to me at a better rate. They can do a pair of the OA80s for $26 postage if I use USPS rather than UPS which is what their site defaults to. That's sounding more attractive. Winding toroidals is not too bad once you have the details of the wind set out and the O-core makes it easier to keep it neat and stress free. It's a nice little project even if done just for the interest, and if it works well it should be a useful unit to have.
I do take your point about what is causing the difference JMF. The H&K has a 1W/5W/18W power soak built in so you can easily get it to the sweet spot then trim the volume with that. I have really ruled out damping factor as an issue now thanks to Sreten's great advice which pointed out that I was viewing it wrongly from the start.
I am also trying out other speaker options, I haven't mentally locked myself into the matching Tx option. I have someone lending me a 16ohm Greenback to go in there but I still need another 16ohm unit to try it out fully, it's only the 25W unit and it'll be underpowered on its own in the long term. I'll come up with something though, this has to be sorted or I'll go mad.
Turner Audio Transformer Winding
This guy knows his onions! Trouble is it takes some time to get through the stuff he's talking about but it's a Godsend and I'll be digesting it soon.
For this application, if I'm going to wind one I'll use a good quality core suited to audio frequency range work Andrew. For anyone interested in this sort of stuff, I would recommend having a look at this range:
Silicon Steel Toroidal O-Cores
O-core is the Holy Grail of this sort of stuff. Don't forget that this is wide range GOSS material in the best shape around. They offer a range of sizes and all incredibly cheap. The OA80 which is a 96-182VA unit, comes out at $14 but the postage to the UK puts the price up $123, not viable no matter how good they are!
I emailed Alpha-Core about the postage cost and got a reply back with an offer to ship their GOSS Toroidal O-core units to me at a better rate. They can do a pair of the OA80s for $26 postage if I use USPS rather than UPS which is what their site defaults to. That's sounding more attractive. Winding toroidals is not too bad once you have the details of the wind set out and the O-core makes it easier to keep it neat and stress free. It's a nice little project even if done just for the interest, and if it works well it should be a useful unit to have.
I do take your point about what is causing the difference JMF. The H&K has a 1W/5W/18W power soak built in so you can easily get it to the sweet spot then trim the volume with that. I have really ruled out damping factor as an issue now thanks to Sreten's great advice which pointed out that I was viewing it wrongly from the start.
I am also trying out other speaker options, I haven't mentally locked myself into the matching Tx option. I have someone lending me a 16ohm Greenback to go in there but I still need another 16ohm unit to try it out fully, it's only the 25W unit and it'll be underpowered on its own in the long term. I'll come up with something though, this has to be sorted or I'll go mad.
The lower impedance load gives worse damping factor.
But the lower impedance load might get a lot more power from your amp, and the distortion characteristics may be very different; that distortion can also cause a shift in perceived frequency balance. The exact results depend on the exact tubes, power supply, and output transformer used.
Yes, having drivers in series can cause the unique impedance curve anomalies, how the two drivers differ, to become exaggerated. For your application, where you were considering 4 drivers, it's common practice to make two pairs, with each pair of drivers in parallel. That averages those anomalies, and the two pairs will match better when you then put the two pairs in series. At least that's the theory, though there are many amplifier manufacturers who do the opposite in order to increase some unusual distortions in musical instruments.
I also suggest that the sound of two drivers is always different than one, even in the same volume of enclosure space. The bass couples, the treble beams. I highly recommend an angled baffle board for musical instruments. I prefer the "V" baffle board with half the driver pointing left a bit and half pointing right a bit. The effect is very noticeable, and there is a manufacturer who makes beveled mounting rings, to point the drivers in slightly different directions; it really works.
Some people think I'm nuts for bothering, but I have had very good results from output transformer upgrades. The result often fits within the original space, unlike adding another matching transformer. Using an additional transformer will degrade performance and double the iron weight; IMHO you're much better off replacing the output transformer with the one you really want. Purchasing an Edcor costs less than my time is worth.
On the other hand, thinking about using two transformers: I've always thought there might be some performance potential in single-ended amps from using a first transformer with massive iron and primary winding to handle the DC but conservative turns ratio followed by a toroidal transformer to do the actual step-down. For a push-pull it is again possible to customize the first transformer for handling the opposed currents in the primary and cancelling magnetic fields, especially for a class-A push/pull, and then the step-down could be toroidal. Two transformers, each customized for the particular task, might out-perform one and also be more efficient. But adding an additional transformer to an existing amp isn't as likely to improve things.
Just one tinkerer's thoughts. And my disclaimer: I'm no EE, I'm just a hobbyist.
But the lower impedance load might get a lot more power from your amp, and the distortion characteristics may be very different; that distortion can also cause a shift in perceived frequency balance. The exact results depend on the exact tubes, power supply, and output transformer used.
Yes, having drivers in series can cause the unique impedance curve anomalies, how the two drivers differ, to become exaggerated. For your application, where you were considering 4 drivers, it's common practice to make two pairs, with each pair of drivers in parallel. That averages those anomalies, and the two pairs will match better when you then put the two pairs in series. At least that's the theory, though there are many amplifier manufacturers who do the opposite in order to increase some unusual distortions in musical instruments.
I also suggest that the sound of two drivers is always different than one, even in the same volume of enclosure space. The bass couples, the treble beams. I highly recommend an angled baffle board for musical instruments. I prefer the "V" baffle board with half the driver pointing left a bit and half pointing right a bit. The effect is very noticeable, and there is a manufacturer who makes beveled mounting rings, to point the drivers in slightly different directions; it really works.
Some people think I'm nuts for bothering, but I have had very good results from output transformer upgrades. The result often fits within the original space, unlike adding another matching transformer. Using an additional transformer will degrade performance and double the iron weight; IMHO you're much better off replacing the output transformer with the one you really want. Purchasing an Edcor costs less than my time is worth.
On the other hand, thinking about using two transformers: I've always thought there might be some performance potential in single-ended amps from using a first transformer with massive iron and primary winding to handle the DC but conservative turns ratio followed by a toroidal transformer to do the actual step-down. For a push-pull it is again possible to customize the first transformer for handling the opposed currents in the primary and cancelling magnetic fields, especially for a class-A push/pull, and then the step-down could be toroidal. Two transformers, each customized for the particular task, might out-perform one and also be more efficient. But adding an additional transformer to an existing amp isn't as likely to improve things.
Just one tinkerer's thoughts. And my disclaimer: I'm no EE, I'm just a hobbyist.
Last edited:
- Status
- Not open for further replies.