"Guys, can you point me to some direction how to use mains toroid with SE amp, i am about to make SE amp with GU-50.
"
What operating point will the GU-50 be at, what plate load will it need. What output power are you expecting.
Given that the GU-50 can deliver high output power and that you will need a toroidal of at least 4-6 times that power, you may well need a huge toroidal. If this is the case the high frequency response will probably suffer badly.
Shoog
"
What operating point will the GU-50 be at, what plate load will it need. What output power are you expecting.
Given that the GU-50 can deliver high output power and that you will need a toroidal of at least 4-6 times that power, you may well need a huge toroidal. If this is the case the high frequency response will probably suffer badly.
Shoog
Hello Shoog, I would like to ask you a few questions since I know you have experience with toroidal OPTs.
From various sources, I read everything and its opposite.
- toroidals are bad because of high capacitance, big secondaries windings (wire diameter) that cause high-frequency losses.
- toroidals have very wide frequency response up to 100kHz, with no resonances in the audio band. This is in strong contrast.
- toroidals have low primary inductance. So big VA ratings are needed, and low driving impedance.
- toroidals are ideal in low frequency amplification in multiamplified systems. I think we're at the opposite ends here...
- toroidals can tollerate some small imbalanced DC current without any effect. So bias servos for PP amps using toroids are not needed, one just has to check the bias once a month or so.
- toroidals can never tolerate even a microampere of DC current without extreme losses in performance.
I think the truth lies in between, what about your experience?
I must add that I haven't made measurements yet, and I'm particulary oriented to PP using of toroidals, not parafeed.
From various sources, I read everything and its opposite.
- toroidals are bad because of high capacitance, big secondaries windings (wire diameter) that cause high-frequency losses.
- toroidals have very wide frequency response up to 100kHz, with no resonances in the audio band. This is in strong contrast.
- toroidals have low primary inductance. So big VA ratings are needed, and low driving impedance.
- toroidals are ideal in low frequency amplification in multiamplified systems. I think we're at the opposite ends here...
- toroidals can tollerate some small imbalanced DC current without any effect. So bias servos for PP amps using toroids are not needed, one just has to check the bias once a month or so.
- toroidals can never tolerate even a microampere of DC current without extreme losses in performance.
I think the truth lies in between, what about your experience?
I must add that I haven't made measurements yet, and I'm particulary oriented to PP using of toroidals, not parafeed.
I would like to build something like this:Shoog said:
http://www.tubeland.de/images/gu50sc.gif
Or this:
http://www.jogis-roehrenbude.de/Leserbriefe/SvenB-GU50/schaltunggross.gif
I am pretty new with tubes, i made some solid state amps, but nothing too big or complicated, so i would like to made something proven
Hi Giaime,
You ask all the relevant questions.
My experience so far with toroidals as output have been in two applications. The main one is in a 807 parafeed amp with the 110V tap used for ultralinear. They are about 120VA with 18V secondaries. I unwound two thirds of the secondaries to get the ratio I wanted. There is plate to plate feedback which keeps the output impedence well down. This allows you to drive the low primary inductance to low frequencies and I get response down to 20hz or lower. Using a high output impedence will produce worse results. The high frequency response is excellent and there is no roll off in the upper frequencies. A small amount of DC is tollerated because the UL arrangement works well, the 807 does draw unusually low screen current though. Overall in the parafeed arrangement they have worked extremely well.
I have also used them in a SE parafeed headphone amp using ECL82's with Plate to Plate feedback, but without UL. Again they have worked well, but the low frequency response isn't so good.
My conclusions would be;
To get low frequency response go for 10x the output power in VA rating.
The direction in which the windings are connected can have a big effect on the response. Connecting the primary in one direction produced low frequency roll off and a high frequency dip.
I have not found any issues with ringing. If there are unterminated windings ringing will become an issue.
Not all toroidals will work well, with high frequency roll off been the main victim.
It is early days with my experiements in toroidal PP's. So far I can report that the same toroidals work reasonably well. I am using 6080's which suffer from poor balance. The toroidals produce good high frequency response. In the lower frequencies they are satisfactory, but imbalance produces badly distorted low frequency waveforms which is especially noticable with square waves. Balancing eliminated this. This was achieved with seperate adjustable cathode resistors. If you do decide not to use a bias servo, seperate adjustable cathode resistors will produce better results.
I also am using toroidals as interstage phase splitters. This is where they have been most disapointing. Low frequency has been good, high frequency need a lot of current to drive the inter winding capacitance. I put this down to the 1:1+1 ratio. Step down ratios seem to work a lot better.
I would conclude by saying that it seems that all toroidals are not made equal. I have been lucky in having a good batch of toroidals which have performed really well.
I hope that helps, if you want clarification on any of those points - please ask.
Shoog
You ask all the relevant questions.
My experience so far with toroidals as output have been in two applications. The main one is in a 807 parafeed amp with the 110V tap used for ultralinear. They are about 120VA with 18V secondaries. I unwound two thirds of the secondaries to get the ratio I wanted. There is plate to plate feedback which keeps the output impedence well down. This allows you to drive the low primary inductance to low frequencies and I get response down to 20hz or lower. Using a high output impedence will produce worse results. The high frequency response is excellent and there is no roll off in the upper frequencies. A small amount of DC is tollerated because the UL arrangement works well, the 807 does draw unusually low screen current though. Overall in the parafeed arrangement they have worked extremely well.
I have also used them in a SE parafeed headphone amp using ECL82's with Plate to Plate feedback, but without UL. Again they have worked well, but the low frequency response isn't so good.
My conclusions would be;
To get low frequency response go for 10x the output power in VA rating.
The direction in which the windings are connected can have a big effect on the response. Connecting the primary in one direction produced low frequency roll off and a high frequency dip.
I have not found any issues with ringing. If there are unterminated windings ringing will become an issue.
Not all toroidals will work well, with high frequency roll off been the main victim.
It is early days with my experiements in toroidal PP's. So far I can report that the same toroidals work reasonably well. I am using 6080's which suffer from poor balance. The toroidals produce good high frequency response. In the lower frequencies they are satisfactory, but imbalance produces badly distorted low frequency waveforms which is especially noticable with square waves. Balancing eliminated this. This was achieved with seperate adjustable cathode resistors. If you do decide not to use a bias servo, seperate adjustable cathode resistors will produce better results.
I also am using toroidals as interstage phase splitters. This is where they have been most disapointing. Low frequency has been good, high frequency need a lot of current to drive the inter winding capacitance. I put this down to the 1:1+1 ratio. Step down ratios seem to work a lot better.
I would conclude by saying that it seems that all toroidals are not made equal. I have been lucky in having a good batch of toroidals which have performed really well.
I hope that helps, if you want clarification on any of those points - please ask.
Shoog
Thank you very much Shoog, you have answered all my questions.
I'm planning to use toroids in a PP amp with EL36 in triode mode, as I understand they have a fairly low internal resistance, so I could be able to drive the transformers properly. Please check plate curves and typical characteristics (thank you Tom!!!):
http://www.tubes.mynetcologne.de/roehren/daten/el36pentode_as_triode_v2.pdf
They will use fixed bias with a trimmer for each tube to balance currents.
I'm planning to use toroids in a PP amp with EL36 in triode mode, as I understand they have a fairly low internal resistance, so I could be able to drive the transformers properly. Please check plate curves and typical characteristics (thank you Tom!!!):
http://www.tubes.mynetcologne.de/roehren/daten/el36pentode_as_triode_v2.pdf
They will use fixed bias with a trimmer for each tube to balance currents.
As i mentioned earlier, for low frequency Power, the VA rating is pretty much meaningless and going to a higher VA core will not increase the amount of voltage you can place across the primary.
I have also noted a direct relationship between primary inductance and VA rating and the smaller the core, the higher the inductance. This is with only maybe 10 samples or so, so i wouldn't consider it proof but rather a pattern.
If you consider a transformer with a fixed ratio from Full load to unloaded primary current (VA/line voltage)/ magnetizing current, as the VA rating goes down so must the no load current to keep the ratio the same. which requires an increase in inductance.
a simple test for this would simply be to measure the no load primary current of a few transformers and use 2piFL to detrmine the primary inductance. The transformer with the lowest current should have the most inductance.
dave
The formula for inductance of wound cores (N turns):
L = u0 *(Urelative) * (N*N) * (Core_Area) / (Core_length)
Voltage rating of winding (N turns):
V = 2*Pi*N*(Core_Area)*(Bmax)*(freq.)
If one doubles all linear dimensions of the core, core area quadruples, so the original N turns can be reduced by 4x for the same voltage rating.
Core length also doubles, so new inductance L' now becomes:
L' = u0 * (Urelative) * (N/4 *N/4) *(Core_Area *4) / (Core_length*2)
L' = L/8
Conclusion: Use a smaller xfmr for more inductance or use one with a higher voltage rating.
Don
L = u0 *(Urelative) * (N*N) * (Core_Area) / (Core_length)
Voltage rating of winding (N turns):
V = 2*Pi*N*(Core_Area)*(Bmax)*(freq.)
If one doubles all linear dimensions of the core, core area quadruples, so the original N turns can be reduced by 4x for the same voltage rating.
Core length also doubles, so new inductance L' now becomes:
L' = u0 * (Urelative) * (N/4 *N/4) *(Core_Area *4) / (Core_length*2)
L' = L/8
Conclusion: Use a smaller xfmr for more inductance or use one with a higher voltage rating.
Don
Choke in the anode would be better. Finding the right choke to set the bias would difficult, or you would need supplementary fixed bias to set it right. Also you would need to bypass the choke for AC.
I have got my 807 parafeed setup with Microwave Oven Transformers as the anode choke load. Works well - but not as good as a CCS.
Shoog
I have got my 807 parafeed setup with Microwave Oven Transformers as the anode choke load. Works well - but not as good as a CCS.
Shoog
hi guys
Just a hint
A variac-type of transformer can make a real good pp or
parafed output transformer for 100Vrms across the usually
0-250V winding.
You have to wind the secondaries by hand on top but it has the advantage that you can "interleave" windings easely. Best suitable size is a variac with a wirethickness below 0.6mm diam.
0.5 will get you beyond the audioband without undue losses.
Just a hint
A variac-type of transformer can make a real good pp or
parafed output transformer for 100Vrms across the usually
0-250V winding.
You have to wind the secondaries by hand on top but it has the advantage that you can "interleave" windings easely. Best suitable size is a variac with a wirethickness below 0.6mm diam.
0.5 will get you beyond the audioband without undue losses.
I am looking for a cheaper parafeed OPT and reading through this thread, I might be able to get away with using a toroidal power transformer.
I am doing a parafeed 12b4 as power tube and 6j5 for driver stage. I will be using a 40H 30ma plate choke for the 12b4, B+ of around 250V cathode resistor of around 1.2K.
I am thinking of using a toroidal transformer for the parafeed output. The toroidal's primary is 0-230 and two secondaries of 0-6V. I computed the primary Z at around 5.8K at 4ohm secondary. The toroidal power ratingis 30VA.
What do you think guys?
Thanks.
I am doing a parafeed 12b4 as power tube and 6j5 for driver stage. I will be using a 40H 30ma plate choke for the 12b4, B+ of around 250V cathode resistor of around 1.2K.
I am thinking of using a toroidal transformer for the parafeed output. The toroidal's primary is 0-230 and two secondaries of 0-6V. I computed the primary Z at around 5.8K at 4ohm secondary. The toroidal power ratingis 30VA.
What do you think guys?
Thanks.
Tony said:hi alexG,
you can get this 110/220volt pri. to 12volts sec. 500mA transformer, cost less than 100 pesos(about U$2) as parafeed output traffo.
use the 110volt winding and 12volt, it will give you a 9.1 turns ratio, so for a 600ohm sec load, reflected impedance would be around 50k,
Tony,
I will be loading the secondary with either 4 or 6 ohm load (speaker load), with 600 ohm load, would that be headphones?
Thanks.
You need to work out the winding ratio, and then the impedance ratio before you can start to work out which tubes might work for you.
Because you have only got a relatively low ratio, you will have to get imaginative about suitable valves for the application. Some of the hi-current low-voltage valves should be your starting point. Look at the PL36,PL509, 6080, or any of the sweep output tubes. The good news is that all of these are cheap and cheerful. All of the more typical output tubes are well out of the question - which is a good thing to my mind.
If you could get the 6080 to work and you used a dual triode and SS rectification, you could get a two valve stereo amp. This approach also makes the design of CCS plate loads slightly easier. The 6080 works excellently up to 100mA down at 100V.
I saw a design for a PL36 cathode follower which might suit these transformers. In this application you don't need a plate load and the Constant Current Sink at the cathode can be a simple LM317.
Shoog
Because you have only got a relatively low ratio, you will have to get imaginative about suitable valves for the application. Some of the hi-current low-voltage valves should be your starting point. Look at the PL36,PL509, 6080, or any of the sweep output tubes. The good news is that all of these are cheap and cheerful. All of the more typical output tubes are well out of the question - which is a good thing to my mind.
If you could get the 6080 to work and you used a dual triode and SS rectification, you could get a two valve stereo amp. This approach also makes the design of CCS plate loads slightly easier. The 6080 works excellently up to 100mA down at 100V.
I saw a design for a PL36 cathode follower which might suit these transformers. In this application you don't need a plate load and the Constant Current Sink at the cathode can be a simple LM317.
Shoog
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.