Voltage across amplifier boards.
Guys whats the voltage range these boards and trannies can take?
I see other builders and articles going between 22v-32V. Is there an ideal voltage?
I realize the more voltage the bigger the heat sinks (which i am making sure is BIG and can take heat)
Also now i'm using chokes my transformer doesn't have to be a massive 800Va?
Can stay at 600va, or just to be safe make it big? Does a big trafo at 600va plus create more magnetic noise?
Also considering dual bridge rectifiers in the design, any tips?
Guys whats the voltage range these boards and trannies can take?
I see other builders and articles going between 22v-32V. Is there an ideal voltage?
I realize the more voltage the bigger the heat sinks (which i am making sure is BIG and can take heat)
Also now i'm using chokes my transformer doesn't have to be a massive 800Va?
Can stay at 600va, or just to be safe make it big? Does a big trafo at 600va plus create more magnetic noise?
Also considering dual bridge rectifiers in the design, any tips?
Hiraga himself recommended this solution and demonstrate on a FFT analysis the diminish of harmonics. I will find the page and put a link to it.
The supplementary noise can refer only to switching noise of the diodes, and for that I suggested snubbers.
However, since calculating or at least experimental determination of snubber values is a bit complicate, I also mentioned that using coils in the filter will reduce much of the switching noise. It also help to bypass rectifying diodes by C =10-22nF/250V MKP (Wima are ok).
Transformers having not 2 separated windings cannot be used for 2 bridges rectification.
http://www.tnt-audio.com/jpg/psu4.jpg
LE - the next explanation from TNTAudio is not true : "... this allows for much better channel separation, since each supply line is independent, and is therefore much less likely to transmit signal from one channel to the other".
For it to be true should be at least separate rectification for each channel if not separate trafo.
Don't take me wrong, I just said what the best solution is, I am not insisting you to do it this way.
In my oppinion the right voltage for the Hiraga 30W is depending on how much power you want to get (Hiraga 40W, 50W, 65W, etc.); the limitation is first given by the radiators you have. Ok, a 65W Hiraga is overkill for the transistors, but with a cryogenic system... 😀
About the trafo - well, bigger is better, but in real life the difference in money vs quality is quite big in my opinion. This for what I saw about R-core prices. I would say that a 600-650VA trafo should be enough.
If you are going for E+I type trafo, a soft start impose itself.
The supplementary noise can refer only to switching noise of the diodes, and for that I suggested snubbers.
However, since calculating or at least experimental determination of snubber values is a bit complicate, I also mentioned that using coils in the filter will reduce much of the switching noise. It also help to bypass rectifying diodes by C =10-22nF/250V MKP (Wima are ok).
Transformers having not 2 separated windings cannot be used for 2 bridges rectification.
http://www.tnt-audio.com/jpg/psu4.jpg
LE - the next explanation from TNTAudio is not true : "... this allows for much better channel separation, since each supply line is independent, and is therefore much less likely to transmit signal from one channel to the other".
For it to be true should be at least separate rectification for each channel if not separate trafo.
Don't take me wrong, I just said what the best solution is, I am not insisting you to do it this way.
In my oppinion the right voltage for the Hiraga 30W is depending on how much power you want to get (Hiraga 40W, 50W, 65W, etc.); the limitation is first given by the radiators you have. Ok, a 65W Hiraga is overkill for the transistors, but with a cryogenic system... 😀
About the trafo - well, bigger is better, but in real life the difference in money vs quality is quite big in my opinion. This for what I saw about R-core prices. I would say that a 600-650VA trafo should be enough.
If you are going for E+I type trafo, a soft start impose itself.
Last edited:
In the previous message I wrote by mistake "radiators" instead of "heatsinks". Sorry.
About trafo voltage.
You must understand that any trafo gives a voltage for 0A and a lower voltage for a certain current.
In case you need - let's say 3,5A (to cover the bias of 1,6-1,7A per channel) you should know how much the trafo voltage will fall at that current. The bigger the trafo, the lower the fall will be.
If you don't know if a trafo quoted for 2x20Vac deliver this voltage for 0A or for 1,5A or for 6A, you must ask the vendor or the producer. Usually the quoted trafos voltage is for the nominal current, so at a lower one the voltage is higher.
For a Hiraga 30W minimum +/-24V is needed, and if you use series chokes, voltage before them should be 3 V higher => +/- 27V (at 3,5A) must be obtained from the PS.
If you use Schottky diodes for rectifying that means the trafo must deliver minimum 2x18,6-2x19Vac at 3,5A.
If classic diodes are used, the voltage must be increase by aprox. 0,5V.
Obviously there is no problem to use a 2x20Vac trafo (at 3,5A).
The extra power the amp deliver with 2x26Vdc instead of 2x24Vdc, it is no a problem if the heatsinks are big enough.
And if necessary you can reduce the bias so the temp. decrease under 70C. (Of course this is undesirable because as everybody wrote, the amp sound best at 1,5-1,68A)
About trafo voltage.
You must understand that any trafo gives a voltage for 0A and a lower voltage for a certain current.
In case you need - let's say 3,5A (to cover the bias of 1,6-1,7A per channel) you should know how much the trafo voltage will fall at that current. The bigger the trafo, the lower the fall will be.
If you don't know if a trafo quoted for 2x20Vac deliver this voltage for 0A or for 1,5A or for 6A, you must ask the vendor or the producer. Usually the quoted trafos voltage is for the nominal current, so at a lower one the voltage is higher.
For a Hiraga 30W minimum +/-24V is needed, and if you use series chokes, voltage before them should be 3 V higher => +/- 27V (at 3,5A) must be obtained from the PS.
If you use Schottky diodes for rectifying that means the trafo must deliver minimum 2x18,6-2x19Vac at 3,5A.
If classic diodes are used, the voltage must be increase by aprox. 0,5V.
Obviously there is no problem to use a 2x20Vac trafo (at 3,5A).
The extra power the amp deliver with 2x26Vdc instead of 2x24Vdc, it is no a problem if the heatsinks are big enough.
And if necessary you can reduce the bias so the temp. decrease under 70C. (Of course this is undesirable because as everybody wrote, the amp sound best at 1,5-1,68A)
Last edited:
Hello,
Did start typing than internet went down so lost my reply.
So just a few lines now.
The right 500/600 VA rcore/double C core with properly designed ( use bleeder to assure current flow) choke input will sound way better than a 800VA without a choke!
This will probably not need a soft start! just dont make the first cap to big. I am using a 500VA R core with ll2733 choke input more than 1A load 10000µF first cap and 2 A slow blow fuse.
So if you can get a 600 VA r core ( 600va is a core Shilcar is selling) 320 is a bit small and 1000Va is BIG and might need a soft start continue looking for that 600VA one!
Make a test circuit with the right load.
Make the circuit work around 24 volts and 1,5A. Why reduce its lifetime by overheating it?
POST some pictures of the test power supply before switching on. Make a light bulb tester.
Will be back after your R core has arrived!
greetings, Eduard
Did start typing than internet went down so lost my reply.
So just a few lines now.
The right 500/600 VA rcore/double C core with properly designed ( use bleeder to assure current flow) choke input will sound way better than a 800VA without a choke!
This will probably not need a soft start! just dont make the first cap to big. I am using a 500VA R core with ll2733 choke input more than 1A load 10000µF first cap and 2 A slow blow fuse.
So if you can get a 600 VA r core ( 600va is a core Shilcar is selling) 320 is a bit small and 1000Va is BIG and might need a soft start continue looking for that 600VA one!
Make a test circuit with the right load.
Make the circuit work around 24 volts and 1,5A. Why reduce its lifetime by overheating it?
POST some pictures of the test power supply before switching on. Make a light bulb tester.
Will be back after your R core has arrived!
greetings, Eduard
Eduard is insisting in the benefits of using chokes for filtering.
I can't agree more.
However, a soft start in not to avoid, even for a 500VA trafo, because of its many advantages (you can use smaller diodes, it prolong the first caps life, eliminates saturation moments for the trafo core and eliminates the light dimm at start-up). The condition is to use a very good big relay, and eventually clean its contacts every few years.
I can't agree more.
However, a soft start in not to avoid, even for a 500VA trafo, because of its many advantages (you can use smaller diodes, it prolong the first caps life, eliminates saturation moments for the trafo core and eliminates the light dimm at start-up). The condition is to use a very good big relay, and eventually clean its contacts every few years.
Last edited:
The core parts of r-cores are preassembled before winding the coils.
R-core coils are round; that is economic in terms of winding speed, but winding r-cores require dedicated winding equipment.
The preassembled r-core has almost no airgap, that's why they can have quite some inrush current.
I said "can" because this also depends on core excitation.
A major benefit of c-cores is that air gapping can be applied easily between the two halves of the core; this way inrush current is not an issue.
By the way, proper use of toroidal transformers is not so difficult: in case of unpleasant inrush current you can apply NTC resistors in the primary. When I am right most manufacturerers of toroidal transformers advice to use NTC resistors for high power toroids.
IMO however a good quality c-core transformer, neatly wound and vacuum impregnated, is superior to any "off the shelve" toroid.
Hello,
Not just chokes but choke input.
Why so few do it? Because they dont know!
Smaller diodes? maybe because if you have a power failure for a very short time. The caps could be empty but the relais still connected so then you might have a to big current destroying the diodes. Diodes with the right current rating will not cost a lot more.
My r cores dont buzz at switch on so i assume the exiting current is pretty low.
I had a resistor in series with the primary winding to reduce the voltage to the circuit to about 12? volts so power consumption would be very low but caps were kept charged. When listening i just bridged this resistor with a high quality switch. It was like a kind of standby.
Bye bye
Not just chokes but choke input.
Why so few do it? Because they dont know!
Smaller diodes? maybe because if you have a power failure for a very short time. The caps could be empty but the relais still connected so then you might have a to big current destroying the diodes. Diodes with the right current rating will not cost a lot more.
My r cores dont buzz at switch on so i assume the exiting current is pretty low.
I had a resistor in series with the primary winding to reduce the voltage to the circuit to about 12? volts so power consumption would be very low but caps were kept charged. When listening i just bridged this resistor with a high quality switch. It was like a kind of standby.
Bye bye
Shambala,
When you are located in the NE of Romania, the town of Iasi is likely not far from your place (you might even live there).
In the past I had some really good quality toroidal transformers wound by Petra Transformers in Iasi.
On request they supplied the transformers with a static screen, and core excitation was specified to be max 1 or 1.2 T at 230 volt primary. These are really good transformers.
A HiFi firm here in the Netherlands still use them extensively in their own-made (expensive) equipment.
When you are located in the NE of Romania, the town of Iasi is likely not far from your place (you might even live there).
In the past I had some really good quality toroidal transformers wound by Petra Transformers in Iasi.
On request they supplied the transformers with a static screen, and core excitation was specified to be max 1 or 1.2 T at 230 volt primary. These are really good transformers.
A HiFi firm here in the Netherlands still use them extensively in their own-made (expensive) equipment.
Agree.
Much in DIY audio is based on "believe" and "tradition".
Toroidal power supply transformers are the norm (also in commercial HiFi by the way).
Their potential drawbacks are often overlooked (toroids huh so it will be good....).
Eduard your power supply already benefits from choke input: it makes the task for the power supply transformer much easier.
I once had trouble with a HV supply for a tube amp; rebuilding it to choke input completely removed the noise from the power supply transformer as the current spikes from the rectification were not reflected back anymore.
Eduard, you are describing a manual soft-start - the most audiophile solution. Just love it 😉
Pieter, I do live in Iasi and have a many years long cooperation with Petra Toroid, for special design trafos. Hope I was not a big pain in the *** of the technical manager; spent many hours calculating together some of those trafos.
In the pic can be seen two big special trafo from Petra, in a passive power conditioner.
Pieter, I do live in Iasi and have a many years long cooperation with Petra Toroid, for special design trafos. Hope I was not a big pain in the *** of the technical manager; spent many hours calculating together some of those trafos.
In the pic can be seen two big special trafo from Petra, in a passive power conditioner.
Attachments
Last edited:
Hello,
Call it what you like but you need to get the value of the resistor right!! I remember destroying a 25 watt resistor in an aluminium housing. The resistorelement inside the aluminum '' tube '' can go nowhere except going left or right and it did with a big bang. If the value is to low the amp will still use close to its nominal current and the charging of thje caps will not be limited enough. If i remember well it was a 15 ohm that i used with the 500 VA EI core.
It was replaced by a higher one 220ohm ? but not sure about that.
In the end i did use the power connector to switch it on. It has been on for a decade and just toggle the switch when i needed it. Most gear '' dies '' during swtich on so using this trick prevents that and saves loads on your electricity bill. Slow starting the charging of the caps is nice when they are expensive as the ones sold in Paris.
When using a regulator circuit you cannot use this trick for that part because it needs a minimum voltage drop to get it working. BUT for the output stage where the current is it is nice.
IF i am right this amp is still working after 35 years.
Greetings, Eduard
Ps Sure choke input gives transformer rectifier and caps an easier life. Sounds nicer, more spacious and especially more powerful
Hey, don't dramatise, chosen the corect R value and power is childplay.
For 500VA you need aprox. 22-23ohm at minimum 50W for a 3-5sec. delay. If you make a manual soft start put 100-200W resistor.
In the very first moment power on R is very big (>2000W) but is decreasing every millisecond up to 230W for a 2x30W Hiraga with 1,6A bias.
The replacement of your blown 15ohm probably was a 22ohm, an 220ohm takes a extremely long time to charge the capacitors.
Vishalk, about the trafo voltage.
I remembered that you choose the pcb with the Kubota stab. for the first stage. What is it written on the doc. they came with, what is the minimum voltage necessary ? It may be necessary to be +/- 26V or +/-28V
Last edited:
I recently build two Q-Watt mono amplifiers using SMPS. The Q-Watt is a class ab amplifier. The results are excellent and the amp works like a charm.
There are many advantages to SMPS's such as lower weight, smaller footprint, cheaper, more stable (against changing load), no hum, etc. The SMPS I used is the SMPS500R. It is a 500 watts power supply and it is available in different variations (dual or singel supply, different voltages). I'm very pleased with them.
There is thread here on the Q-Watt amplifier and there pictures of the various builds: http://www.diyaudio.com/forums/chip-amps/301100-my-q-watt-project.html
Hello,
The purpose of the resistor is make the ac output of the transformer lower. If you 15 ohm and dont take care to switch it off it will blow.
If my memory serves me well dont use 15 ohm. It will go boom pretty soon.
Wait for the experts to confirm lol.
Greetings, Eduard
Delange, congratulations for your nice built.
As well known there are advantages and disadvantages in using SMPS for audio amps. As there are for a linear PS, of course. Everyone is free to chose the way considered best for him.
Personally I prefer classic PS, even for a class A amp (that is more suited for a SMPS than a class AB, because its constant consumption).
Eduard, no, the purpose of the resistor [of the soft start, be it manual or automatic] is not to make the ac output of the transformer lower, that is a consequence. The purpose is to limit the inrush current.
As well known there are advantages and disadvantages in using SMPS for audio amps. As there are for a linear PS, of course. Everyone is free to chose the way considered best for him.
Personally I prefer classic PS, even for a class A amp (that is more suited for a SMPS than a class AB, because its constant consumption).
Eduard, no, the purpose of the resistor [of the soft start, be it manual or automatic] is not to make the ac output of the transformer lower, that is a consequence. The purpose is to limit the inrush current.
Last edited:
There is no reason to do that... only if you already have them and there are no money spent.
If you bypass them by a relay is it better to use R, not Th.
If you bypass them by a relay is it better to use R, not Th.
Why is resistor better? If relay fails to operate for any reason nothing is going to go bang...