Hi,
It's work and well !
It's work and well !
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
More details :
power supply : 28,1 V for one rail and 27,9 for the other (speakers load)
with 8 Ohm resistor 3 mV at output
after 15 mn working temp is 37 °C at top of heatsink
At listening (with poor speakers), it's clear, good bass, voices are very nice, all is very pleasant and musical..
Another question, I think it's possible to remove C5 (input cap) with LM's ?
Phil.
power supply : 28,1 V for one rail and 27,9 for the other (speakers load)
with 8 Ohm resistor 3 mV at output
after 15 mn working temp is 37 °C at top of heatsink
At listening (with poor speakers), it's clear, good bass, voices are very nice, all is very pleasant and musical..
Another question, I think it's possible to remove C5 (input cap) with LM's ?
Phil.
No, do not remove the input cap.
It does two quite separate jobs.
It blocks accidental DC entering from a Source.
It defines the LF pass frequency to ensure the amplifier does not get fed with a signal that it cannot properly handle.
It does two quite separate jobs.
It blocks accidental DC entering from a Source.
It defines the LF pass frequency to ensure the amplifier does not get fed with a signal that it cannot properly handle.
Hi,
I just finished a diy "box"..
Phil.
I just finished a diy "box"..
Phil.
An externally hosted image should be here but it was not working when we last tested it.
Hi Steve,
Yes and it's works every days..
Since I removed input cap (it's a risk) but I think (hope) my dac is secure with free of dc.
Next step, build a new LM with chipamp kit (dual mono) for mid and high, and convert this amp for bass.
Phil.
Yes and it's works every days..
Since I removed input cap (it's a risk) but I think (hope) my dac is secure with free of dc.
Next step, build a new LM with chipamp kit (dual mono) for mid and high, and convert this amp for bass.
Phil.
Upgrade 2015/02
Hi all,
Some changes :
C1 and C2 are 470 uF (0,1 uF //) instead of 100 uF
R6 (input resistor) 330R (Takman carbon) instead of 1K
remove C5 (input cap)
I want to remove R1 (output resistor 0,1R 5w), what is the risk with instability ?
Add a zobel will be difficult..
Phil.
Hi all,
Some changes :
C1 and C2 are 470 uF (0,1 uF //) instead of 100 uF
R6 (input resistor) 330R (Takman carbon) instead of 1K
remove C5 (input cap)
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
I want to remove R1 (output resistor 0,1R 5w), what is the risk with instability ?
Add a zobel will be difficult..
Phil.
Hi,
I remove R1, high frequencies sounding is more "clean"..
DC offset with R1 (0,1R) at 8 Ohm load is 0,2 mV, and 2,3 mV without R1..
Phil.
I remove R1, high frequencies sounding is more "clean"..
DC offset with R1 (0,1R) at 8 Ohm load is 0,2 mV, and 2,3 mV without R1..
An externally hosted image should be here but it was not working when we last tested it.
Phil.
R1 should have a parallel output inductor. 0u5H to 1u5H
see post9.
I prefer to have R1||L1 off the PCB and put it in the cable to the output terminals. Away from the PCB and away from any steel/iron metal work.
see post9.
I prefer to have R1||L1 off the PCB and put it in the cable to the output terminals. Away from the PCB and away from any steel/iron metal work.
Hi Andrew,
Yes, you are right !
Pict of this "resistor" :
Phil.
Yes, you are right !
Pict of this "resistor" :
An externally hosted image should be here but it was not working when we last tested it.
Phil.
Hi,
Adding RL (4,7R), nothing changes at sounding, maybe I prefer without !
Phil.
An externally hosted image should be here but it was not working when we last tested it.
Adding RL (4,7R), nothing changes at sounding, maybe I prefer without !
Phil.
nothing changes.
Comparing what to what?
What is the DCR of the coil? It looks quite high, using thin wire and many turns, maybe getting close to your 0r1 you originally had.
Use >1mm diameter copper and use a bigger diameter with many fewer turns to get the DCR down.
7 Turns of 1.6mm copper around a 15mm bobbin should have such a low DCR, that it is equivalent to the length of wire it replaces. i.e. no added resistance.
BTW,
the spkr return should be close coupled, or twisted, with the spkr flow.
The input signal flow should be twisted with the signal return.
Eliminate the LOOP AREA at the terminals !
You have two sets of aerial systems creating avoidable interference.
You PSU wiring looks like it is braided. Twist it as a triplet right up to very close to the PCB.
Comparing what to what?
What is the DCR of the coil? It looks quite high, using thin wire and many turns, maybe getting close to your 0r1 you originally had.
Use >1mm diameter copper and use a bigger diameter with many fewer turns to get the DCR down.
7 Turns of 1.6mm copper around a 15mm bobbin should have such a low DCR, that it is equivalent to the length of wire it replaces. i.e. no added resistance.
BTW,
the spkr return should be close coupled, or twisted, with the spkr flow.
The input signal flow should be twisted with the signal return.
Eliminate the LOOP AREA at the terminals !
You have two sets of aerial systems creating avoidable interference.
You PSU wiring looks like it is braided. Twist it as a triplet right up to very close to the PCB.
Last edited:
nothing changes ?
Between without R1 (0,1R) serial output, and RL..
R is 4,7R, coil is 14 turns awg 18 diameter 12mm.
Maybe cause my speakers have Zobel network ?
It's a test setup, to see what's happen by changing values and componants, so it is not definitive, which explains why wires are not twisted.
Thank's.
Next step, change psu by chipamp with snubber.
Phil.
Between without R1 (0,1R) serial output, and RL..
R is 4,7R, coil is 14 turns awg 18 diameter 12mm.
Maybe cause my speakers have Zobel network ?
It's a test setup, to see what's happen by changing values and componants, so it is not definitive, which explains why wires are not twisted.
Thank's.
Next step, change psu by chipamp with snubber.
Phil.
Test setup or not there is no reason to not twist the output wires.
The inputs should be coax cables.
Next time don't use so much solder on the PC board.
The inputs should be coax cables.
Next time don't use so much solder on the PC board.
Of course, but what difference at sounding ?
What kind of coax cable, an example ?
You are right, it was my first trial with LM..
Phil.
It's 'good engineering practice' to twist all multi-conductor cables, 99% of the time there is no difference, but it's that 1% that's the problem.
Any handy, useable coax will do.
Any handy, useable coax will do.
I find it much easier to use twisted pair, rather than coax, for line level connections.
This is especially so inside an amplifier chassis where solid core hook up wire (0.6mm diam) makes for easy routing that "stays" where you want it.
CAT5/CAT5e off cuts, which is slightly smaller in diameter (0.5mm), also do well for internal twisted connections.
I only use coax for ready made interconnects and for lower signal levels than line level, eg. TT cartridge & microphone.
This is especially so inside an amplifier chassis where solid core hook up wire (0.6mm diam) makes for easy routing that "stays" where you want it.
CAT5/CAT5e off cuts, which is slightly smaller in diameter (0.5mm), also do well for internal twisted connections.
I only use coax for ready made interconnects and for lower signal levels than line level, eg. TT cartridge & microphone.
- Status
- Not open for further replies.