Hi,
i want to built a new preamp and i woult like to know what's the best topology for better sound. For input: jfet or bipolair, single ended or complimentary, cascode or differential or differential cascode same think for VAS and for output stage: jfet, bipolair or mosfet? Thank you! Maxpou
i want to built a new preamp and i woult like to know what's the best topology for better sound. For input: jfet or bipolair, single ended or complimentary, cascode or differential or differential cascode same think for VAS and for output stage: jfet, bipolair or mosfet? Thank you! Maxpou
You may wish to experiment with a FET version of the Aikido amp.
I ve got a tremendous improvement when I replaced a NE5534
with a Aikido config. with n channel FETS and MOSFETS. However
I used 2N5196 which is a dual n channel FET which is very accurately matched. The results may not be as good with unmatched FETs.
I ve got a tremendous improvement when I replaced a NE5534
with a Aikido config. with n channel FETS and MOSFETS. However
I used 2N5196 which is a dual n channel FET which is very accurately matched. The results may not be as good with unmatched FETs.
Before choosing your topology first consider your requirements. You need to consider...
>The output impedance of the source equipment (i.e. the equipment before the preamplifier)
>The input impedance of the target equipment (i.e. the equipment after the preamplifier)
>The impedance of the connecting cables
>Any likely heat issues (for example, is it going to be in a confined space?)
>Any likely vibration issues (for example, is it going to be near a loudspeaker?)
>The desired maximum gain (or very possibly the desired attenuation) - for this you need to consider the output level of the source equipment and the gain of the target equipment
>The likelyhood of a clean utility mains power supply
>The likelyhood of local electro-magnetic interfearance, both radiated and conducted
Also, when planning your project keep the following in mind...
> Ground distribution has a major influence on your circuit performance. Ground is not ground... instead regard it as 1. a safety point, 2. a power supply reference point, 3. a signal return (in terms of 'conventional current'). In other words spend some time on getting it right.
> Most 'real' music signals spend a lot of time at low level. At low level the cross-over distortion of a class AB stage can sometimes be a significant contribution to 'decreased sound quality'. As far as I know 99.9% of op-amp output stages work in class AB.
> Single ended solid state stages may be susceptible to thermal drift if the ambient temperature change is more than a few degrees.
> Potentiometers can be a source of significant channel-to-channel imbalance which varies with volume setting, and this can seriously detract from the overall system performance.
... and now you can start to choose your circuit design.
Personally I suggest you consider either single-ended CCS-loaded FET or vacuum tubes in an Aikido type of topology (see tubecad.com).
Good luck with it : )
>The output impedance of the source equipment (i.e. the equipment before the preamplifier)
>The input impedance of the target equipment (i.e. the equipment after the preamplifier)
>The impedance of the connecting cables
>Any likely heat issues (for example, is it going to be in a confined space?)
>Any likely vibration issues (for example, is it going to be near a loudspeaker?)
>The desired maximum gain (or very possibly the desired attenuation) - for this you need to consider the output level of the source equipment and the gain of the target equipment
>The likelyhood of a clean utility mains power supply
>The likelyhood of local electro-magnetic interfearance, both radiated and conducted
Also, when planning your project keep the following in mind...
> Ground distribution has a major influence on your circuit performance. Ground is not ground... instead regard it as 1. a safety point, 2. a power supply reference point, 3. a signal return (in terms of 'conventional current'). In other words spend some time on getting it right.
> Most 'real' music signals spend a lot of time at low level. At low level the cross-over distortion of a class AB stage can sometimes be a significant contribution to 'decreased sound quality'. As far as I know 99.9% of op-amp output stages work in class AB.
> Single ended solid state stages may be susceptible to thermal drift if the ambient temperature change is more than a few degrees.
> Potentiometers can be a source of significant channel-to-channel imbalance which varies with volume setting, and this can seriously detract from the overall system performance.
... and now you can start to choose your circuit design.
Personally I suggest you consider either single-ended CCS-loaded FET or vacuum tubes in an Aikido type of topology (see tubecad.com).
Good luck with it : )
Hi,
Thank you for your answer guys, but i forgot to mention that i want to built a solid state preamp. I have a player YBA special (modified with jfet single ended for output and regulated power supply for analog circuit). My power amp is Buzquito (mosquito) of Jean-Marc Plantefeve modified for Vmosfet at output stage ( Zin:47K). I have sulzer/borbely negative or positivepower supply with AN817 for my preamp. i have a good potentiometers alps blue 100K, alps black 30K and noble 20K. The gain of preamp can be 12db. Thank you! Maxpou
Thank you for your answer guys, but i forgot to mention that i want to built a solid state preamp. I have a player YBA special (modified with jfet single ended for output and regulated power supply for analog circuit). My power amp is Buzquito (mosquito) of Jean-Marc Plantefeve modified for Vmosfet at output stage ( Zin:47K). I have sulzer/borbely negative or positivepower supply with AN817 for my preamp. i have a good potentiometers alps blue 100K, alps black 30K and noble 20K. The gain of preamp can be 12db. Thank you! Maxpou
For a preamp schematic I suggest that you could use the circuit posted here by forum member called x-pro (Alex Nikitin):
http://www.diyaudio.com/forums/showthread.php?s=&threadid=109286
It was designed as a headphone amplifier but should make a good line stage. More info here...
http://www.ant-audio.co.uk
http://www.diyaudio.com/forums/showthread.php?s=&threadid=109286
It was designed as a headphone amplifier but should make a good line stage. More info here...
http://www.ant-audio.co.uk
Gordy said:
Gordy, thanks for mentioning my design. That circuit is quite flexible and could make a very good SE preamp stage. It should be possible to reduce power consumption by using a higher value of the load resistor on the output follower - i.e. 330-470 Ohm in place of 68 Ohm. I've done it with my phono stage based upon the same circuit configuration and it gives very nice sound, better than my old reference OBH-8SE. The gain would be reasonable to set about 7-10, to provide best linearity and SNR. Perhaps I should try to make a "preamp version" of my circuit and share it here
.Maxpou, if you would like to try my circuit, there is a couple of notes:
1) you'll need about 25-30 J112 transistors to find a suitable quad for one unit. Thankfully these are very cheap. With BSP129 you may need again about the same number to get required 6 for the circuit. You'll need to get 4 with fairly close matching of drain current, preferably near 100mA at 5V (to measure it you need to be quick as the transistor would heat up and the current would drift a bit. Usually I can match MOSFETS down to 1% on the drain current value).
2) you can replace source resistors with 4 trimmers and do a very precise tunung on distortion - after that you can replace these with appropriate fixed value resistors, as pots have more added noise.
I usually measure a batch of 30-50 J112 or J113 with a single "reference" BSP129, 15V supply and 1K load, adjusting a pot in the source of J112 for lowest distortion at 100 mV 1kHz input, writing down the output voltage level (i.e. the gain), the optimum resistor value and the drain current for each device. After that it is easy to find matching transistors.
The sound quality of this circuit is quite dependant on the quality of the output capacitor. For the preamp you can use a good make 10uF polypropylene capacitor for best results.
Cheers
Alex
Hi M. Nikitin,
thank you very much it's very appreciated however i have questions on your schematic. I would like to know if it's possible for me to replacing the jfet by 2SK170 or K246 and mosfet 2SK214 because i have many devices matched of each in stock.
When you tell me:
The gain would be reasonable to set about 7-10.
you want tell volt or db.
Thank You! Maxpou
thank you very much it's very appreciated however i have questions on your schematic. I would like to know if it's possible for me to replacing the jfet by 2SK170 or K246 and mosfet 2SK214 because i have many devices matched of each in stock.
When you tell me:
The gain would be reasonable to set about 7-10.
you want tell volt or db.
Thank You! Maxpou
maxpou said:
Hi Maxpou,
7-10 times (as I did not put "dB" there) or 14-20 dB. While it should be possible to try 2SK170 in place of J112 (thought I don't know how linear these would be - you need to experiment and see for yourself), to replace BSP129 with 2SK214 would be quite difficult, because it would require additional biasing circuits around the MOSFETs and most likely would worsen the performance. BSP129 is a DMOS - depletion-mode FET, behaving similarly to a large JFET. I chose that particular device for number of reasons, including the sound quality. One of the main advantages of BSP129 is it's very high output impedance, providing very good output linearity. These transistors are not expensive and easily available from Farnell, Digikey etc.
Cheers
Alex
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.