that's ok
I keep working with the Baslers then, have enough of them and also some Abbotts...
non toroids though and I was hoping to make something with a see through round enclosure...
J-P
I keep working with the Baslers then, have enough of them and also some Abbotts...
non toroids though and I was hoping to make something with a see through round enclosure...
J-P
I was in need of a small (20WRMS max) stereo amp with input selector, volume control and headphone outlet that would communicate music for the "computer hours" i.e. 3-4 hours daily. I had a steel enclosure from an abandoned project, so no metal work was needed, a well made 2x14,5V/2A E-I x-former, two LM 1875 and a rainy weekend in front of me. Two bridges with discrete BYV28-100 diodes followed by 100uF, (one PSU common for both channels), inverted topology LM1875, Rin=10K, Rfb=330K, Vin+ tied to grnd., 1000uF//0.1uF per polarity, 1uF across pins 3 & 5, shunt volume control 22k series, 100k carbon Log shunt potentiometer. No “audiophile grade” parts.
. Twisted solid core telephone wires for everything, careful routing and earthing. The result was an amp with stunning speed and clarity, guts, very good sound balance and despite the necessary long internal cabling (40cm signal in, 50cm signal out, 40cm PSU power, 50cm 230VAC ), no hum, buzz or other noise. For the record, here are some measurements (Both channels driven, 50 Ohm 0 Hz-100KHz signal generator, 8 Ohm dummy load, 20 Mhz dual analog Scope, digital multimeter. Ambient temp 18 d C., rel. hum. 60%, heatsinks 45 +/- 5 d C, case vibration +/- 2 nm, 0.00001g vertical, +/- 3pm, 0.000001g horizontal) :
1. PSU
DC: +/- 20Vdc.
Ripple (100Hz) measured at the pins (#3,5)of the LMs:
Both LMs Undriven : 450mVp/p.
Both LMs at Full Output (14 WRMS): 4Vp/p
Note: -DC had approx 10% higher ripple than +DC
PSU DC modulation by audio signal (Both Channels driven to Full Output):
Signal freq. (Hz) Modulation amplitude (Vp/p)
50 8
100 8
500 1.5
1000 0.5
2. LMs Output:
AC:
0V input signal : Output (noise modulated by 100Hz signal) 2mVp/p
2Vp/p input signal: Output 30Vp/p just before (1mV) clipping, i.e. 14W RMS/8Ohm
Freq. Response:
Flat +/- 1 db 5Hz-100KHz.(the limit of my signal generator).
The amp goes much higher actually. One day I observed a high freq. signal at the amp. output, something btn. 50 and 60 MHz. I thought that it started oscillating. After some *&$@# I was lucky enough to find out that what was oscillating was actually the signal source and this high freq. signal was faithfully amplified by the LM1875. So, band limiting the input is not a bad idea.
Input/Output Phase Shift:
8 Hz: 10 d.
70KHz: 15 d.
15Hz-30Khz: 4 d. max.
DC offset (mV RMS):
No input signal applied. Input jack was loaded with various dummy resistances, representing the driving source output resistance. For each such dummy source, three measurements were taken. One with volume control 100% (full travel), one at 50% of the travel, and one at 0%.:
Dummy Source: 0 Ohms
Volume position LH Amp RH Amp
100% travel -50 -60
50% travel -54 -64
0% travel -100 -133
Dummy Source: 1 KOhm
Volume position LH Amp RH Amp
100% travel -54 -58
50% travel -54 -64
0% travel -107 -134
Dummy Source: 10 KOhm
Volume position LH Amp RH Amp
100% travel -40 -45
50% travel -45 -52
0% travel -109 -135
Dummy Source: 100 KOhm
Volume position LH Amp RH Amp
100% travel -29 -35
50% travel -35 -44
0% travel -107 -134
Dummy Source: 1 MOhm
Volume position LH Amp RH Amp
100% travel -26 -31
50% travel -32 -41
0% travel -108 -134
Dummy Source: Open
Volume position LH Amp RH Amp
100% travel -28 -29
50% travel -35 -38
0% travel -107 -133
The maximum DC offset of 134mV equals 2.3mW RMS into 8 Ohms and occurs at 0% volume setting. Normally the volume will be between 25% and 50%, where the DC offset induced output is 0.2-0.4 mW . Not a big deal. The variation of the DC offset due to the input loading though indicates, that in principle, a buffer is needed. If it will have detrimental effects on the sound, is a question.
Regards
George
. Twisted solid core telephone wires for everything, careful routing and earthing. The result was an amp with stunning speed and clarity, guts, very good sound balance and despite the necessary long internal cabling (40cm signal in, 50cm signal out, 40cm PSU power, 50cm 230VAC ), no hum, buzz or other noise. For the record, here are some measurements (Both channels driven, 50 Ohm 0 Hz-100KHz signal generator, 8 Ohm dummy load, 20 Mhz dual analog Scope, digital multimeter. Ambient temp 18 d C., rel. hum. 60%, heatsinks 45 +/- 5 d C, case vibration +/- 2 nm, 0.00001g vertical, +/- 3pm, 0.000001g horizontal) :
1. PSU
DC: +/- 20Vdc.
Ripple (100Hz) measured at the pins (#3,5)of the LMs:
Both LMs Undriven : 450mVp/p.
Both LMs at Full Output (14 WRMS): 4Vp/p
Note: -DC had approx 10% higher ripple than +DC
PSU DC modulation by audio signal (Both Channels driven to Full Output):
Signal freq. (Hz) Modulation amplitude (Vp/p)
50 8
100 8
500 1.5
1000 0.5
2. LMs Output:
AC:
0V input signal : Output (noise modulated by 100Hz signal) 2mVp/p
2Vp/p input signal: Output 30Vp/p just before (1mV) clipping, i.e. 14W RMS/8Ohm
Freq. Response:
Flat +/- 1 db 5Hz-100KHz.(the limit of my signal generator).
The amp goes much higher actually. One day I observed a high freq. signal at the amp. output, something btn. 50 and 60 MHz. I thought that it started oscillating. After some *&$@# I was lucky enough to find out that what was oscillating was actually the signal source and this high freq. signal was faithfully amplified by the LM1875. So, band limiting the input is not a bad idea.
Input/Output Phase Shift:
8 Hz: 10 d.
70KHz: 15 d.
15Hz-30Khz: 4 d. max.
DC offset (mV RMS):
No input signal applied. Input jack was loaded with various dummy resistances, representing the driving source output resistance. For each such dummy source, three measurements were taken. One with volume control 100% (full travel), one at 50% of the travel, and one at 0%.:
Dummy Source: 0 Ohms
Volume position LH Amp RH Amp
100% travel -50 -60
50% travel -54 -64
0% travel -100 -133
Dummy Source: 1 KOhm
Volume position LH Amp RH Amp
100% travel -54 -58
50% travel -54 -64
0% travel -107 -134
Dummy Source: 10 KOhm
Volume position LH Amp RH Amp
100% travel -40 -45
50% travel -45 -52
0% travel -109 -135
Dummy Source: 100 KOhm
Volume position LH Amp RH Amp
100% travel -29 -35
50% travel -35 -44
0% travel -107 -134
Dummy Source: 1 MOhm
Volume position LH Amp RH Amp
100% travel -26 -31
50% travel -32 -41
0% travel -108 -134
Dummy Source: Open
Volume position LH Amp RH Amp
100% travel -28 -29
50% travel -35 -38
0% travel -107 -133
The maximum DC offset of 134mV equals 2.3mW RMS into 8 Ohms and occurs at 0% volume setting. Normally the volume will be between 25% and 50%, where the DC offset induced output is 0.2-0.4 mW . Not a big deal. The variation of the DC offset due to the input loading though indicates, that in principle, a buffer is needed. If it will have detrimental effects on the sound, is a question.
Regards
George
George,
Thanks for the excellent description. It sounds like a great amp.
If you are in a tweaking mode, can I propose a resistor of 10k from +in to GND. I think you'll get much better DC at 0% travel (if that is 0 volume, 10k res. connected to gnd).
Or 22k-47k between -In and GND and +in 10k-100k to gnd.
Greg
Thanks for the excellent description. It sounds like a great amp.
If you are in a tweaking mode, can I propose a resistor of 10k from +in to GND. I think you'll get much better DC at 0% travel (if that is 0 volume, 10k res. connected to gnd).
Or 22k-47k between -In and GND and +in 10k-100k to gnd.
Greg
Greg thanks for the comments.
In fact i tried all these you propose, one by one and combined. Variations of +/- 10mV was the result. So i ended with the simplest schematic, the one that i reported. I think that the offset is caused to large extent to the long signal route from input jacks to volume control to LMs. This i can not change as i have to settle to this chassis. As i wrote before, it is not much of a problem. On three past projects (LM1875 non inverted, TDA7294 inverted, LM3886 inverted) all of which had much smaller signal routing, the DC offset was less than about 60mV worst case.
Regards
George
In fact i tried all these you propose, one by one and combined. Variations of +/- 10mV was the result. So i ended with the simplest schematic, the one that i reported. I think that the offset is caused to large extent to the long signal route from input jacks to volume control to LMs. This i can not change as i have to settle to this chassis. As i wrote before, it is not much of a problem. On three past projects (LM1875 non inverted, TDA7294 inverted, LM3886 inverted) all of which had much smaller signal routing, the DC offset was less than about 60mV worst case.
Regards
George
I have a pair of Yamaha NS200's and love them muchly, however I need some advice re: the amplifier to do them justice. Please help...
Thanks
Thanks
tolka said:
I really like the way they sound with the amp that I build. Much more resolution and freedom of the sound, compare to the Yamaha amp that I have. Highs and mids are great and base is plenty also. I'd say, put together one and see how it sounds to you.
Greg
Re: Re: I am ready to start a new
And the flux will reduce too. That is a good thing.
Peter Daniel said:
And the flux will reduce too. That is a good thing.
????
schematic please
and also, i have 4 transformers 110 volts in, 3 wire out 12.5V-0-12.5V , 8 A
how should I go around using them at their best for the GC....
J-P.
(beware you guys I have a lot of transformers here!)
schematic please
and also, i have 4 transformers 110 volts in, 3 wire out 12.5V-0-12.5V , 8 A
how should I go around using them at their best for the GC....
J-P.
(beware you guys I have a lot of transformers here!)
I have another serious question, Gregc or someone else?
how can I use these transformers for a gainclone?
I have 4 of them.
do I only need 2 of them or all 4 for a left and right channel?
can I use 2 of them and make leads to the left and right amps out of the bridge?
How do I make them suitable for a gc?
Thanks
Jean-Pierre
how can I use these transformers for a gainclone?
I have 4 of them.
do I only need 2 of them or all 4 for a left and right channel?
can I use 2 of them and make leads to the left and right amps out of the bridge?
How do I make them suitable for a gc?
Thanks
Jean-Pierre
Attachments
Re: I have another serious question, Gregc or someone else?
You need one transformer and one Bridge per rail. So two transformers and two bridges will give you single power supply for the two channels (L and R). So one traf for +V and one for –V. With the voltages you measured you are very close to the max. allowable voltage for LM 3578 (it’s 84v/2=42V per rail absolute max.) You’ll have 27v*1.41=38V and with 10% fluctuation you’ll get to 42V. Too close in my opinion but it should work.
If you don't need more that 30W (my case) you can yuse 1 traf per chanel +/- 13*1.41=18Vdc. I suggest do that first and if you need more pwer than just reconnect the pwer supply.
uvodee said:
You need one transformer and one Bridge per rail. So two transformers and two bridges will give you single power supply for the two channels (L and R). So one traf for +V and one for –V. With the voltages you measured you are very close to the max. allowable voltage for LM 3578 (it’s 84v/2=42V per rail absolute max.) You’ll have 27v*1.41=38V and with 10% fluctuation you’ll get to 42V. Too close in my opinion but it should work.
If you don't need more that 30W (my case) you can yuse 1 traf per chanel +/- 13*1.41=18Vdc. I suggest do that first and if you need more pwer than just reconnect the pwer supply.
i was thinking
Gregc, I was thinking more in this direction.
these transformers (abbottt technologies. inc. ) are pretty hefty puppies (mil. grade) and quite the works!
as they are 8 A, so 200 VA, I was thinking of using only one (1) for both channels, it is to say using 2 wires to one bridge and parallel another 2 wires to the others bridge and these bridges would then lead to their L or right channel.
Don't you think the transformer would be powerfull enough for this. I think I could live with around 20 to 25 watts, I mainly ilsten to classical anyway.
Jean-Pierre.
ps I don't need to economise on these transformers, I have 4 anyway and can get many more for $15 each.
Gregc, I was thinking more in this direction.
these transformers (abbottt technologies. inc. ) are pretty hefty puppies (mil. grade) and quite the works!
as they are 8 A, so 200 VA, I was thinking of using only one (1) for both channels, it is to say using 2 wires to one bridge and parallel another 2 wires to the others bridge and these bridges would then lead to their L or right channel.
Don't you think the transformer would be powerfull enough for this. I think I could live with around 20 to 25 watts, I mainly ilsten to classical anyway.
Jean-Pierre.
ps I don't need to economise on these transformers, I have 4 anyway and can get many more for $15 each.
Re: i was thinking
Jean-Pierre,
In my setup I use 1 220VA traf per chanel of about 40-50W. Mind you I use no more than 5-10 W when listening to music.
In your case you can use only one transforme for the two chanels (as you mentioned). Use two bridges and create a star GND in the PS section. Run separate set of 3 wires (+,- and GND) to each Amp to minimize interference through the PS wires. I think it'll sound great.
My only concirn was the voltage. 13VAC is fine for 20W/8Ohms. per chanel. If you have 4Ohm speackers the power goes up to 40W. On the other hand the 27VAC would be really too close to the max. voltage. So I'd do what you suggested.
uvodee said:
Jean-Pierre,
In my setup I use 1 220VA traf per chanel of about 40-50W. Mind you I use no more than 5-10 W when listening to music.
In your case you can use only one transforme for the two chanels (as you mentioned). Use two bridges and create a star GND in the PS section. Run separate set of 3 wires (+,- and GND) to each Amp to minimize interference through the PS wires. I think it'll sound great.
My only concirn was the voltage. 13VAC is fine for 20W/8Ohms. per chanel. If you have 4Ohm speackers the power goes up to 40W. On the other hand the 27VAC would be really too close to the max. voltage. So I'd do what you suggested.
great
I am going to inspect some damaged property (water flooding) and when I come back this evening, I am going to get this new project going.
That will be my 3rd gc!
Thanks
J-P.
I am going to inspect some damaged property (water flooding) and when I come back this evening, I am going to get this new project going.
That will be my 3rd gc!
Thanks
J-P.
fedde said:
And it will blow.😱
Two bridges with a CT trafo?😕
Jean-Pierre, I'm sure that you can easily mod that trafo for independent secondaries (4 wires).
Just separate the two wires that are joined together in the centre.😉
- Status
- Not open for further replies.