Hi everyone!
A quick question,
Will this 30VA power transformer, present a reflected load somewhere near 7K/10K, hooking 8 ohm on the secondary?
Rapid Electronics - Electrical & Power
I mean with the primary series connected and parallel secondary., just as this guy does:
http://www.triodeguy.com/Triodeguy%20PDF%20files/6AS7%20CURRENT%20BALANCED%20PP%20AMP%20SCH.pdf
push pull use, of course, and CCSed (sink)
Thanks
J.
A quick question,
Will this 30VA power transformer, present a reflected load somewhere near 7K/10K, hooking 8 ohm on the secondary?
Rapid Electronics - Electrical & Power
I mean with the primary series connected and parallel secondary., just as this guy does:
http://www.triodeguy.com/Triodeguy%20PDF%20files/6AS7%20CURRENT%20BALANCED%20PP%20AMP%20SCH.pdf
push pull use, of course, and CCSed (sink)
Thanks
J.
Thanks!, so I was(quasi) on the ball park, In fact, a 230V/7V unit will be better , but hard to find..
It seems that the measures implemented by Triodeguy, solve the DC unbalance issue.
Forumer Shoog had a lot of experience with toroids, and apparently managed to solve the problem too.
I was thinking on a cheap self splitting circuit, not a true high quality unit.
Thanks again
J.
It seems that the measures implemented by Triodeguy, solve the DC unbalance issue.
Forumer Shoog had a lot of experience with toroids, and apparently managed to solve the problem too.
I was thinking on a cheap self splitting circuit, not a true high quality unit.
Thanks again
J.
Just a tip better than a 60Hz one.
Out of joke, it's a matter of perspective, just suppose you're going to a place where there is no chace to found real OPT's no EDCOR, no Hammond, ...and you still want to have fun with tubes...
Besides some forumers report excellent results with those toroids (at least the Talema units) So, for a cheap 8W self split circuit, let give it a try...
Out of joke, it's a matter of perspective, just suppose you're going to a place where there is no chace to found real OPT's no EDCOR, no Hammond, ...and you still want to have fun with tubes...
Besides some forumers report excellent results with those toroids (at least the Talema units) So, for a cheap 8W self split circuit, let give it a try...
Last edited:
OPTs have always been a problem for me too in the UK. I wish Edcor was here, but they are not and home grown stuff is expensive.
First I would say that you need a much bigger toroid than the power suggests, because the frequency you want to get down to is < 50Hz. So you need more iron!
Secondly you are not bound by the output impedance of the tubes because you can apply feedback to lower it. Not global feedback (to avoid any stability issues), but local between the anode of the output tube to the cathode of its driver tube. Paralleling up tubes is also an option.
So if you have a 12k output impedance from your tube and want to drive a 8R load you need sqrt(12000/8) = 39:1, but if you add enough feedback to reduce the impedance by a factor of 4 you'll need sqrt(3000/8) = 19:1 instead - into 12V transformer territory.
First I would say that you need a much bigger toroid than the power suggests, because the frequency you want to get down to is < 50Hz. So you need more iron!
Secondly you are not bound by the output impedance of the tubes because you can apply feedback to lower it. Not global feedback (to avoid any stability issues), but local between the anode of the output tube to the cathode of its driver tube. Paralleling up tubes is also an option.
So if you have a 12k output impedance from your tube and want to drive a 8R load you need sqrt(12000/8) = 39:1, but if you add enough feedback to reduce the impedance by a factor of 4 you'll need sqrt(3000/8) = 19:1 instead - into 12V transformer territory.
Yes I think I am, how can I find out a little more about how to distinguish the two? TIA!
Apart from turns ratio you need to be sure:
1) Core area is enough for flux density at desired output power.
2) Primary inductance is enough for desired output power/lowest frequency.
3) It should be capable to sustain 5 - 7 mA of output current imbalance (quite complex calc, requires permeability curves).
From what I can see on the web site you posted link to you could probably get only few W. Simply put, it is too small. From the impedance you post I can guess you are looking to build EL84/6P14P based PP amp, aren't you? Then, you need to select core for 15 - 25W / 20 Hz.
LinuksGuru, it is a 6AQ5/6V6. I think you're right, probably iI need to look at >50VA units. I completely forgot that these guys are in the 3 / 4W territory , while I'm looking for 7 / 8W.
http://greygum.net/sbench/analogengineer/7695mat.pdf
7695 Matrix Amp
http://greygum.net/sbench/analogengineer/7695mat.pdf
7695 Matrix Amp
Power toroidals make excellent output transformers (or at least the ones I have used).
The few measurements of primary inductance I have seen are well above what is needed (in the hundreds H as reported by Steve Bench in the design posted by Mosquito), though not all transformers are equal in this regard so its a suck it and see situation.
I can guarantee that the performance will equal or better most designed EI OT with response from 10hz out to 60khz before significant tail off. They will perform much better at low signal levels.
I have used them in a 807 PP amp and they work, but they work best in low output impedance designs. As such the best results can be had with tubes like the 6080. The main reason for this is that when you have to try to create 8K reflected loads the turns ratio's leave very few secondary windings on the core. Still the 807 amp work.
I would always design with at least 4x the expected wattage to VA rating to avoid core saturation.
DC imbalance is the real issue here, and needs to be held to below 2mA. That is not that difficult to achieve, but requires some silicone based autobiasing system.
Anyone who tells you they will not work - has never tried it.
Shoog
The few measurements of primary inductance I have seen are well above what is needed (in the hundreds H as reported by Steve Bench in the design posted by Mosquito), though not all transformers are equal in this regard so its a suck it and see situation.
I can guarantee that the performance will equal or better most designed EI OT with response from 10hz out to 60khz before significant tail off. They will perform much better at low signal levels.
I have used them in a 807 PP amp and they work, but they work best in low output impedance designs. As such the best results can be had with tubes like the 6080. The main reason for this is that when you have to try to create 8K reflected loads the turns ratio's leave very few secondary windings on the core. Still the 807 amp work.
I would always design with at least 4x the expected wattage to VA rating to avoid core saturation.
DC imbalance is the real issue here, and needs to be held to below 2mA. That is not that difficult to achieve, but requires some silicone based autobiasing system.
Anyone who tells you they will not work - has never tried it.
Shoog
Flux Density in Gauss B = 10^8 * sqrt(Power * Primary Impedance) / (4.44 * Area * Stacking Factor * Primary # of Turns * Lowest Frequency)
Flux density should not exceed 15000 G
Stacking factor 0.92 .. 0.95
IMHO you should look at 250 - 300VA toroid at least.
You can quite simply determine primary number of turns on factory unit - wind 10 - 20 turns, connect let's say 100 Ohm resistor, feed transformer from 220V and measure voltage on your small secondary.
Yes that's the way I do it.
I bypass the big caps cathode to cathode with some quality film caps.
Works extremely well.
I built a self splitter that way using EL86's but it kept blowing the Chip based LM317's CCS's. This suggests that the cathodes were spiking high voltages at switch on which was killing them. Adding a zener just below their voltage rating should sort that one out. Still waits to be resolved in my amp though.
I tend to agree that something at least 100VA for even a small amp is probably a good idea.
Shoog
I bypass the big caps cathode to cathode with some quality film caps.
Works extremely well.
I built a self splitter that way using EL86's but it kept blowing the Chip based LM317's CCS's. This suggests that the cathodes were spiking high voltages at switch on which was killing them. Adding a zener just below their voltage rating should sort that one out. Still waits to be resolved in my amp though.
I tend to agree that something at least 100VA for even a small amp is probably a good idea.
Shoog
Yes, I saw that Volt Second proposed a CCS with zeners in the right places, to prevent this issue.
I am thinking of building a small toroid OPT tube amplifier using a bridged tube output, so the toroid just hangs in the middle of two class A/A2 totem-poles. I suspect for best results a Circlotron may be the answer but I quite fancied a dual SRPP arrangement with dual triodes, for simplicity.
I know I could use a capacitor, but I fancied eliminating those!
I guess you can use toroids in push-pull too? You just need an indicator of DC current balance.
I know I could use a capacitor, but I fancied eliminating those!
I guess you can use toroids in push-pull too? You just need an indicator of DC current balance.
I see. I'd modify that myself;- if the wiper loses contact you get full conduction, I'd arrange it so that a dodgy wiper makes it more negative to switch off the tubes..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.