Hello Sjoerd,
In the original publications about these amps you can read they compared more than 20 resistors and the tantal ones were the best. So i would not just throw in loads of metal film resistors!The suggestions about changing this 1K8 was given to me in Paris. I dont know if it was a customer or a staff member of the shop.
Changing the drivers probably also will need changing these small caps on the circuit board.
I think a real BIG change could be made by changing the power supply into choke input!!!
And not using a toriodal transformer. I would go for an R core.
BUT you will surely need an transformer with a higher voltage output. With my dddac where there is about 950 mA current i am getting around 16 volts DC after the input choke LL2733 ( common mode connected) using schottky diodes and a 21 volt transformer. So you probably will need a 36 volt transformer.
Greetings, Eduard
In the original publications about these amps you can read they compared more than 20 resistors and the tantal ones were the best. So i would not just throw in loads of metal film resistors!The suggestions about changing this 1K8 was given to me in Paris. I dont know if it was a customer or a staff member of the shop.
Changing the drivers probably also will need changing these small caps on the circuit board.
I think a real BIG change could be made by changing the power supply into choke input!!!
And not using a toriodal transformer. I would go for an R core.
BUT you will surely need an transformer with a higher voltage output. With my dddac where there is about 950 mA current i am getting around 16 volts DC after the input choke LL2733 ( common mode connected) using schottky diodes and a 21 volt transformer. So you probably will need a 36 volt transformer.
Greetings, Eduard
Hi Sjoerd,
I have seen your thread "what sounds better than Hiraga classe a", and i forgot to answer. I have readed allot of Hiragas impressions and comparisons, wich said that one of the few amps that beats the the Hiraga 20 or 30 watters, and le monster,excepting the valves amps, at that time, was "the super nemesis compense" by Pierre Johannet, giving about 10w at 2% dist and 15w at 8%, but this amp was developed differently (to mimic 300b sound and distortion) in a way that these distortions where thermic distorsions, wich i think is very interesting.
The amp is a pure se, but the interesting part is that it has another transistor wich works at the same time with the output transistor, without signall, connected in antiphase with the output trafo,and at the same bias . This interesting ideea its not new in the se amps and eliminates the gap from the output trafo and increases allot the bass response and the needing of biamping as in the first nemesis amp case.
A very good regulated psu that was developed here was used in the nemesis compensee Document sans titre.
The amp was discussed here Document sans titre and here
Document sans titre.
This will be my next project after finishing my speakers but untill then i want to say some things about my experimenting for two years on my super classe a amp modifyied at 38w sinus:
*i also tested 10 types of resistors (carbons,inductive wirewounds, metal oxide, metal film, power metal), for the input and found that the best sound (cleanest, detailed) from ordinary resistors, no exotics,was the vishay dale rn55;
* the best heat dissipation and abit of clearance aded to the sound was paralelling two 3k6 metal film 2w for the 1k8 bias resistor (they have now 45degrees measured) ; i used ordinary carbon and inductive wirewound here with not such good results, subjectivelly speaking;
*tested carbon, two different wirewound producers and metal oxide resistors, but in the end three paralelled mo 1ohm at 3wats resistors for the 0,33position with the best results;
* tested 1ohm@20w and 0,22@30w wirewound resistors for psu filtering and the 1 ohm heats like fire at 1,65A bias and has abit muddy sound, dont know how to explain it, but with 0,22ohm wich runs barelly hot, the sound is everywhere more cleaner, and feels like it has deeper and rounder bass and the sound feels more airy;
*a cleaner and more detailed sound was also given by 1% matching the input pair; i also matched them 1% diagonally and tested and felt no difference overall, only the "all of them matched" variant sounded abit cleaner at maximum level (i use philips bc 550/560c);
*used stiroflex caps and silver mica on the drivers and felt no difference, maybe its in stability, i dont know but i sticked to the silver mica because i buyed them and costed me some money;
* i used five types of ordinary carbon pots, two multiturn russian types, and two multiturns from bourns, and a sealled wirewound from bourns, and found that the best stability and noise had the wirewound (same concluded Hiraga also,and found a japanese cosmos ra12p wirewound the best); i havent tested multiturn vishay foils or other exotics;
*tested a bunch of transistors for the outputs and found that the best was MJ21194,mjl21194, (they are the same, only their capsules differ, but tested them both) and 2sta1492 and their complements;
Things to test:
1. Did anybody try paralelling 10 pcs of 3kohm resistor instead only one 300R from the feedback, or 10*6k8 parallel for the input?
2. A friend of myne who makes valve amps sugested ayrton perry wirewound n.i. resistor for the input and feedback; he said that in his amp this sounded best. Did anyone tested these types of wirewound? I have seen in audionote page these types of resistors. He sayed that it will take almost 4hours to diy winde a resistor like that, so what do you guys say?
Cheers
Sergiu
I have seen your thread "what sounds better than Hiraga classe a", and i forgot to answer. I have readed allot of Hiragas impressions and comparisons, wich said that one of the few amps that beats the the Hiraga 20 or 30 watters, and le monster,excepting the valves amps, at that time, was "the super nemesis compense" by Pierre Johannet, giving about 10w at 2% dist and 15w at 8%, but this amp was developed differently (to mimic 300b sound and distortion) in a way that these distortions where thermic distorsions, wich i think is very interesting.
The amp is a pure se, but the interesting part is that it has another transistor wich works at the same time with the output transistor, without signall, connected in antiphase with the output trafo,and at the same bias . This interesting ideea its not new in the se amps and eliminates the gap from the output trafo and increases allot the bass response and the needing of biamping as in the first nemesis amp case.
A very good regulated psu that was developed here was used in the nemesis compensee Document sans titre.
The amp was discussed here Document sans titre and here
Document sans titre.
This will be my next project after finishing my speakers but untill then i want to say some things about my experimenting for two years on my super classe a amp modifyied at 38w sinus:
*i also tested 10 types of resistors (carbons,inductive wirewounds, metal oxide, metal film, power metal), for the input and found that the best sound (cleanest, detailed) from ordinary resistors, no exotics,was the vishay dale rn55;
* the best heat dissipation and abit of clearance aded to the sound was paralelling two 3k6 metal film 2w for the 1k8 bias resistor (they have now 45degrees measured) ; i used ordinary carbon and inductive wirewound here with not such good results, subjectivelly speaking;
*tested carbon, two different wirewound producers and metal oxide resistors, but in the end three paralelled mo 1ohm at 3wats resistors for the 0,33position with the best results;
* tested 1ohm@20w and 0,22@30w wirewound resistors for psu filtering and the 1 ohm heats like fire at 1,65A bias and has abit muddy sound, dont know how to explain it, but with 0,22ohm wich runs barelly hot, the sound is everywhere more cleaner, and feels like it has deeper and rounder bass and the sound feels more airy;
*a cleaner and more detailed sound was also given by 1% matching the input pair; i also matched them 1% diagonally and tested and felt no difference overall, only the "all of them matched" variant sounded abit cleaner at maximum level (i use philips bc 550/560c);
*used stiroflex caps and silver mica on the drivers and felt no difference, maybe its in stability, i dont know but i sticked to the silver mica because i buyed them and costed me some money;
* i used five types of ordinary carbon pots, two multiturn russian types, and two multiturns from bourns, and a sealled wirewound from bourns, and found that the best stability and noise had the wirewound (same concluded Hiraga also,and found a japanese cosmos ra12p wirewound the best); i havent tested multiturn vishay foils or other exotics;
*tested a bunch of transistors for the outputs and found that the best was MJ21194,mjl21194, (they are the same, only their capsules differ, but tested them both) and 2sta1492 and their complements;
Things to test:
1. Did anybody try paralelling 10 pcs of 3kohm resistor instead only one 300R from the feedback, or 10*6k8 parallel for the input?
2. A friend of myne who makes valve amps sugested ayrton perry wirewound n.i. resistor for the input and feedback; he said that in his amp this sounded best. Did anyone tested these types of wirewound? I have seen in audionote page these types of resistors. He sayed that it will take almost 4hours to diy winde a resistor like that, so what do you guys say?
Cheers
Sergiu
need help is arranging PSU Caps
hi friends,
i'm putting together PSU for Hiraga 30W amp. this is a mono build and I've the following capacitors for each channel:
5X100,000 uF caps
4X10,000 uF caps
what is the best way to wire them? i'll be using 0.3 Ohms , 30W resistor for CRC filter. probably i made a mistake in sourcing odd number of [big]caps for each PSU!?
hi friends,
i'm putting together PSU for Hiraga 30W amp. this is a mono build and I've the following capacitors for each channel:
5X100,000 uF caps
4X10,000 uF caps
what is the best way to wire them? i'll be using 0.3 Ohms , 30W resistor for CRC filter. probably i made a mistake in sourcing odd number of [big]caps for each PSU!?
Last edited:
thanks @eduard for the quick response. planning to change to choke little later. right now still in the process of finalizing the build.
so for each rail:
2X10,000 uF + 0.33 Ohm + 2X100,000 uF correct?
i do not want to waste 1 unit of 100,000 uF. probably i'll try to source 2 more of these .
so for each rail:
2X10,000 uF + 0.33 Ohm + 2X100,000 uF correct?
i do not want to waste 1 unit of 100,000 uF. probably i'll try to source 2 more of these .
Did the test, dont tell me 2 chokes on one core won't work. Test schematic as by OP.
31 V, 14700 uf, ripple voltage (measured with fluke 83) 0,47 V. Load 2 x 10 ohm.
CLC, 4700 u, dual choke, 10000 u, only E iron inserted ripple 0,15 V
E and I inserted, zero/minimum airgap 84 mV ripple
1 paper thickness 68 mV
2 49
3 33
4 32
5 33
7 34
10 40
15 49
Quite educational and useful. And I will use this for a Hiraga power supply with first cap 40000uf en second cap 160000uf so I guess ripple won't be a problem.
31 V, 14700 uf, ripple voltage (measured with fluke 83) 0,47 V. Load 2 x 10 ohm.
CLC, 4700 u, dual choke, 10000 u, only E iron inserted ripple 0,15 V
E and I inserted, zero/minimum airgap 84 mV ripple
1 paper thickness 68 mV
2 49
3 33
4 32
5 33
7 34
10 40
15 49
Quite educational and useful. And I will use this for a Hiraga power supply with first cap 40000uf en second cap 160000uf so I guess ripple won't be a problem.
@Steven thanks for the recommendation. i'll check if i can get the choke made locally.
Hi eduard,
regarding my earlier query of capacitors:
can I arrange the capacitors in the following manner?
Rail 1 : 2X10,000[63V DC] uF + 0.33[30W] Ohm + 3X100,000[50V DC] uF
Rail 2 : 2X10,000[63V DC] uF + 0.33[30W] Ohm + 2X100,000[50V DC] uF+ 10X10,000[63V DC] uF
i've got the 1,00,000 uF caps after much struggle [spent lot of time and money] . so don't feel like wasting it
. manufacturer is asking for a minimum order quantity of 10 units!
Hi eduard,
regarding my earlier query of capacitors:
can I arrange the capacitors in the following manner?
Rail 1 : 2X10,000[63V DC] uF + 0.33[30W] Ohm + 3X100,000[50V DC] uF
Rail 2 : 2X10,000[63V DC] uF + 0.33[30W] Ohm + 2X100,000[50V DC] uF+ 10X10,000[63V DC] uF
i've got the 1,00,000 uF caps after much struggle [spent lot of time and money] . so don't feel like wasting it
. manufacturer is asking for a minimum order quantity of 10 units!
Last edited:
Below is the arrangement for my PSU which i am using for 30 Watt Jean Hiraga. i Am using 8700uf x 10 Caps for Each Channel insted of 22000uf x 3 Caps.
Schematic of Amp & Power Supply.
Below Image is of my PCB layout with Equivalent transistors as Originals are not available locally.
In Below Image you can see 2 Boards, the Bigger one was my First Board to check if the PCB is working or not, Found some Faults with Tracks which later were corrected in 2nd PCB, which is the improved layout. Both are Tested and are Sounding Superb. I can Say i had never heard such a clean Amp.
The Heat Sink is very small for this Amplifier & It gets very Hot. Each PCB is tested one at a time. One cannot touch the hands for More than 4 to 5 Seconds.
All the above PCB's are Made my me right from Itching to Final Assembly. Its a complete DIY.
Sadik
Schematic of Amp & Power Supply.
Below Image is of my PCB layout with Equivalent transistors as Originals are not available locally.
In Below Image you can see 2 Boards, the Bigger one was my First Board to check if the PCB is working or not, Found some Faults with Tracks which later were corrected in 2nd PCB, which is the improved layout. Both are Tested and are Sounding Superb. I can Say i had never heard such a clean Amp.
The Heat Sink is very small for this Amplifier & It gets very Hot. Each PCB is tested one at a time. One cannot touch the hands for More than 4 to 5 Seconds.
All the above PCB's are Made my me right from Itching to Final Assembly. Its a complete DIY.
Sadik
@Steven thanks for the recommendation. i'll check if i can get the choke made locally.
Hi eduard,
regarding my earlier query of capacitors:
can I arrange the capacitors in the following manner?
Rail 1 : 2X10,000[63V DC] uF + 0.33[30W] Ohm + 3X100,000[50V DC] uF
Rail 2 : 2X10,000[63V DC] uF + 0.33[30W] Ohm + 2X100,000[50V DC] uF+ 10X10,000[63V DC] uF
i've got the 1,00,000 uF caps after much struggle [spent lot of time and money] . so don't feel like wasting it
Hello,
The best would be to have the two rails using the same values and looking indentical. Also take care where you connect the wires to the resistors and to the circuit. I did post a drawing copied from the French magazine some time ago.
And of coursew use a real choke like the ones made by Lundahl. If the currents flowing through the positive and the negative side are identical you can use one coil for pos and the other for negative side. I did ask Per Lundahl himself so time ago. Then you will need 2 chokes for two mono amps.
In the past the French did use 200 mH for a 50 watt class A amp and the Nemesis so i did ask if i could use it for my Hiraga. They said sure why not.
Greetings, Eduard
P.s VERY few people are using these Lundahl chokes for amps like Nelson Pass and Hiraga. In december i will take a set of LL2733 for a friend in Vietnam to use it in a Pass design. Usually people use low value chokes like Hammond ( 10mH) or even air core . I think Lundahl ius the one to use
Now that's the Difference when a Mechanical Engineers Puts his leg in Electronics. Just want to know what will be the benefits of using such a heavy Power Supply. I Tested the Kits with the PS i made and is working Outstanding.
Sadik
@eduard thanks , i'll keep it identical.
for choke, i'll try to get hold of one.
@Sadik , yeah i read that on the other forum. if i'm not wrong there are people who have tried this design with SMPS PSU too. but then some also say that power supply makes up half the amp. so i guess you can experiment. anyway you can always beef it up later if required!
for choke, i'll try to get hold of one.
@Sadik , yeah i read that on the other forum. if i'm not wrong there are people who have tried this design with SMPS PSU too. but then some also say that power supply makes up half the amp. so i guess you can experiment. anyway you can always beef it up later if required!
ypjc
my trusted dealer- if it is possible for you.
2SA1943-O - Toshiba Power Transistor -230V -15A
Power-Complementary online at Profusion
my trusted dealer- if it is possible for you.
2SA1943-O - Toshiba Power Transistor -230V -15A
Power-Complementary online at Profusion
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.