Extreme_Boky said:
I was doing a repair on a console once, the studio owner fired up
16 channels direct from the 2"/16T Ampex tape MC, no converters,
and I went "WOW, what is that!"
He had this big grin and said "sounds like hell doesn't it!"
Australian colloquialism, hell = good
Here's something interesting for tape MC lovers.
http://www.cco.caltech.edu/~boyk/351.htm
http://www.cco.caltech.edu/~boyk/rep-int.htm
cheers,
Terry
I have investigated the rejection of the circuit to the power supply. The result is that there is no a difference between feeding leds with 1mA and 5mA.
In an experiment with a power supply of 13v a curling of feeding 0.4Vpp and with a diode LED and resistance to simplify the circuit. The results were these:
1mA attenuation of the curling 60.2db Impedance of Led 10.0
5mA attenuation of the curling 62.0db Impedance of Led 2.0
As it is seen is certain improvement with 5mA. But he is not very great.
The result of the impedance says to us that once the diode reaches its tension (to zener). The impedance is linear not like in zener that it has an impedance characteristic.
I have not observed an improvement evident to attack Q21 with a smaller impedance. A change if but without reaching an evident conclusion.
Therefore the factor noise seems to be most interesting.
In an experiment with a power supply of 13v a curling of feeding 0.4Vpp and with a diode LED and resistance to simplify the circuit. The results were these:
1mA attenuation of the curling 60.2db Impedance of Led 10.0
5mA attenuation of the curling 62.0db Impedance of Led 2.0
As it is seen is certain improvement with 5mA. But he is not very great.
The result of the impedance says to us that once the diode reaches its tension (to zener). The impedance is linear not like in zener that it has an impedance characteristic.
I have not observed an improvement evident to attack Q21 with a smaller impedance. A change if but without reaching an evident conclusion.
Therefore the factor noise seems to be most interesting.
Jocko Homo said:
Hi
True, 2 things happen when increasing the current:
Output impedance of current source lowers
Dynamic impedance of LED lowers
You may want to repeat the measurement by using for example a 10 ohm resistoir in stead of an LED. The you are able to answer Jocko's question
Again, it is only 2 dB, but to understand matters is interesting. And the lowest noise may be more important.....
best
Hello I have prepared a small circuit to make comparisons between different components and to see the situation with less confusion.
The curling of the power supply is of 0.8Vpp. the current that circulates around diode LED will be regulated by RL.
The results for a diode LED were these:
1mA attenuation 30db
2mA attenuation 28.5db
5mA attenuation 18.8db
If we replaced diode LED by zener of 3.9V 0.33W and equaled the conditions. Zener the results they were these:
1mA attenuation 27.8db
5mA attenuation 26.0db
If we replaced diode LED by diodes formed by transistors union BE and equaled the conditions. For transistors union BE the results were these:
1mA attenuation 22.0db
5mA attenuation 21.0db
In the previous experiment a zero were strained to me, pardon.
It surprises the attenuation of the LED, 1mA 30db. All the components have behaved like in the theory. With its corresponding differences.
Conclusion:
For currents of 1mA the LED is better. For greater currents as 5mA or is better more the LED. As our interest is the one of a smaller noise and a better rejection of the power supply our election would be the LED with 1mA. If in the election we looked for a LED of new generation and chosen by I catalogue (High efficiency or hiper luminous) we could even improve this parameter. It is evident that for each necessity, one will have to make its experiments. This single example would be been worth for this circuit.
I believe that already I am going to construct the circuit with the pertinent changes.
The curling of the power supply is of 0.8Vpp. the current that circulates around diode LED will be regulated by RL.
The results for a diode LED were these:
1mA attenuation 30db
2mA attenuation 28.5db
5mA attenuation 18.8db
If we replaced diode LED by zener of 3.9V 0.33W and equaled the conditions. Zener the results they were these:
1mA attenuation 27.8db
5mA attenuation 26.0db
If we replaced diode LED by diodes formed by transistors union BE and equaled the conditions. For transistors union BE the results were these:
1mA attenuation 22.0db
5mA attenuation 21.0db
In the previous experiment a zero were strained to me, pardon.
It surprises the attenuation of the LED, 1mA 30db. All the components have behaved like in the theory. With its corresponding differences.
Conclusion:
For currents of 1mA the LED is better. For greater currents as 5mA or is better more the LED. As our interest is the one of a smaller noise and a better rejection of the power supply our election would be the LED with 1mA. If in the election we looked for a LED of new generation and chosen by I catalogue (High efficiency or hiper luminous) we could even improve this parameter. It is evident that for each necessity, one will have to make its experiments. This single example would be been worth for this circuit.
I believe that already I am going to construct the circuit with the pertinent changes.
Attachments
If we replaced diode LED by diodes 1N4148. For diodes 1N4148 the results were these:
1mA attenuation 20.0db
5mA attenuation 28.9db
Jfets belongs to another category of current sources. And the CCS always to improve the parameters. This experiment was to verify that current he was better for leds. A little has been extended and we have seen interesting things.
Jocko, You can investigate and also communicate your investigations to us. The circuits for this type of experimentation are simple.
Greetings
1mA attenuation 20.0db
5mA attenuation 28.9db
Jfets belongs to another category of current sources. And the CCS always to improve the parameters. This experiment was to verify that current he was better for leds. A little has been extended and we have seen interesting things.
Jocko, You can investigate and also communicate your investigations to us. The circuits for this type of experimentation are simple.
Greetings
hello CraigBuckingham
The circuit already has been solved. That part that your you mention. It introduces a series of advantages in the circuit.
Summarizing the advantages:
1 The circuit provides a rejection to the power supply.
2 The circuit is more immune to the interferences.
3 The signal has less noise like parasitic oscillations.
4 The signal has one better relation of 50% of the half-cycle.
5 The circuit is faster. We have greater bandwidth and less errors of phase.
Greetings
The circuit already has been solved. That part that your you mention. It introduces a series of advantages in the circuit.
Summarizing the advantages:
1 The circuit provides a rejection to the power supply.
2 The circuit is more immune to the interferences.
3 The signal has less noise like parasitic oscillations.
4 The signal has one better relation of 50% of the half-cycle.
5 The circuit is faster. We have greater bandwidth and less errors of phase.
Greetings
Hello Sabas,
I was referring to http://www.diyaudio.com/forums/attachment.php?s=&postid=752551&stamp=1130316600
If they are the results you are getting from this circuit then your simulator is broken unless you are injecting unusual amounts of noise into your power supply rail. A simple RC on the Vcc will usually suffice.
Regards, Craig.
I was referring to http://www.diyaudio.com/forums/attachment.php?s=&postid=752551&stamp=1130316600
If they are the results you are getting from this circuit then your simulator is broken unless you are injecting unusual amounts of noise into your power supply rail. A simple RC on the Vcc will usually suffice.
Regards, Craig.
hello CraigBuckingham
The verified circuit this in hardware. It is not a simulation. The scheme that I put to him is so that you they can see the scheme. The voltages are measured real.
As it can see here Signal
You are right to be able to verify the first point. An experiment with a power supply became with 0.8Vpp of curling. With a good power supply it had not been able to see nothing.
All the points that I have put have compared with experiments hardware. They are not simulations. Ground not to use simulations single if the circuit is very complicated or dangerous.
And the observations are made comparing them with a normal circuit and executed well.
I see that the circuit is rare but it has his advantages.
Regards, Sabas.
The verified circuit this in hardware. It is not a simulation. The scheme that I put to him is so that you they can see the scheme. The voltages are measured real.
As it can see here Signal
You are right to be able to verify the first point. An experiment with a power supply became with 0.8Vpp of curling. With a good power supply it had not been able to see nothing.
All the points that I have put have compared with experiments hardware. They are not simulations. Ground not to use simulations single if the circuit is very complicated or dangerous.
And the observations are made comparing them with a normal circuit and executed well.
I see that the circuit is rare but it has his advantages.
Regards, Sabas.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.