I haven't found any specific Leach Amp topics, so I'm just gonna post here.
I'd like to hear some opinions about the 10/47/180/390 pF capacitors. I'm having a dilemma whether to use silvered mica or polystyrene for these values.
Another dilemma are the output transistors. MJL21193/4 pair or the MJL4281A/4302A?!
My only goal is to archive the best sound quality. I'm not realy interested in technical specifications.
Any opinion is welcome.
I'd like to hear some opinions about the 10/47/180/390 pF capacitors. I'm having a dilemma whether to use silvered mica or polystyrene for these values.
Another dilemma are the output transistors. MJL21193/4 pair or the MJL4281A/4302A?!
My only goal is to archive the best sound quality. I'm not realy interested in technical specifications.
Any opinion is welcome.
I was planning to do his 6 transistor version design, but his PCB has pretty bad size issues. i.e. forcing me to use polyester caps due to pin spacing and such.
So I'm planning to do it with the original Leach amp (components and) values with a PCB of my own. For those low values I'd rather use teflon caps if such values would be available. Both Jens and Dr. Leach are suggesting mica caps. However some people say micas are a bit too "spicy" and not neutral at all. I have no experience with polystyrene. That's why I'm asking which would provide a "better" sound. Most "reviews" are being discussed for different audio applications. I have found nothing about the effect in using such caps in solid state amps.
As for the transistors:
MJE4281A has better SOA
MJE21193 has better gain linearity.
I'm pretty sure both are an excellent choice, however I'd still like to hear some opinions about them.
So I'm planning to do it with the original Leach amp (components and) values with a PCB of my own. For those low values I'd rather use teflon caps if such values would be available. Both Jens and Dr. Leach are suggesting mica caps. However some people say micas are a bit too "spicy" and not neutral at all. I have no experience with polystyrene. That's why I'm asking which would provide a "better" sound. Most "reviews" are being discussed for different audio applications. I have found nothing about the effect in using such caps in solid state amps.
As for the transistors:
MJE4281A has better SOA
MJE21193 has better gain linearity.
I'm pretty sure both are an excellent choice, however I'd still like to hear some opinions about them.
duekfx,
I have two of the Jens 10 transistor amps completely built up and running in my workshop. I used all of the parts exactly as stated on the Jens BOM. The mica caps are in where suggested and so forth. I am running these amps with the bias set 750 ma per rail -- down from one amp. They sound excellent. There was no second guessing on the parts choices since I am not an audiophile and can usually not hear any difference if all of the parts are quality items to begin with. My only real splurge item was the mills resistors on the outputs. I used the 4281/4302 transistors. Rod Elliot who is not known for exotic components raved about these transistors.
This is a great amp. However you decide to build it you will be quite pleased.
The only other good amps I have had to compare it to were my old Mcintosh 2250 and 2200 amps. I sold it when the Leach was finished. The Leach being a better unit.
Tad
I have two of the Jens 10 transistor amps completely built up and running in my workshop. I used all of the parts exactly as stated on the Jens BOM. The mica caps are in where suggested and so forth. I am running these amps with the bias set 750 ma per rail -- down from one amp. They sound excellent. There was no second guessing on the parts choices since I am not an audiophile and can usually not hear any difference if all of the parts are quality items to begin with. My only real splurge item was the mills resistors on the outputs. I used the 4281/4302 transistors. Rod Elliot who is not known for exotic components raved about these transistors.
This is a great amp. However you decide to build it you will be quite pleased.
The only other good amps I have had to compare it to were my old Mcintosh 2250 and 2200 amps. I sold it when the Leach was finished. The Leach being a better unit.
Tad
duekfx said:
Not all Mica caps are comparable/identical, same goes for Polystyrene capacitors.
The purity of Mica from the various global resources varies considerably, as well as the type/color, this determines the quality of the capacitor.
There's Silver Mica caps from CDE/CD, SAN, SAHA, SAMWHA, MORGAN, SUNTAN, to name just a few of the familiar labels.
Not a bleeding idea what to think of Rusky manufactured SM's, you could ask Anatoliy Lisovskiy (the Aikido Hedgehog) Wavebourn for his expert opinion.
The absolute purest/clearest Mica is found in India, but the common type of Mica used in Asian produced Silver Micas is a lesser grade by far.
Same story for polystyrene caps, but worse.
Difference between a cheap axial polystyrene cap, that goes for half a dime overhere, and a top rank one is staggering.
Production of most elite polystyrene cap types has stopped long time ago, the sporadic remaining ones are not exactly a bargain offer.
And then there is ceramic COG.
duekfx said:
If you use polystyrene, use only extended foil type.
If you use mica, use a quality brand like CDE. Cheap ones have reliability issues.
Another alternative is Class 1 NP0/CG0 ceramics. Much cheaper and easier to get. Don't let anyone tell you "never use ceramics". This type of ceramic is a different animal from Class 2 and 3 ceramics that are not good. Also, they should fit well.
You would probably do well with any of these types.
The Leach amp clone I have in my workshop has an 8 volt increased voltage front end from a separate pair of transformers and caps. It is not regulated but it is isolated. I personally can not hear any difference and since I do not have a scope I can not measure any difference. I do think there is a small increase in volume with the separate power supply front end setup.
If I was to do this over again I would not spend the time with the extra power supply for sonic reasons. If I were to rely on bench measurements for my music enjoyment maybe the extra work would be appreciated.
If you are into this for the hobby aspect you might find the extra circuitry as a challenge to your skills and include it. It is all fun so do what you want since you can. The nice thing is however you build your Leach you will have a very nice amplifier, which you constructed, for the rest of your life. I sold off some extremely nice audiophile components after building my Leach and passive pre and have not regretted it once.
Tad
If I was to do this over again I would not spend the time with the extra power supply for sonic reasons. If I were to rely on bench measurements for my music enjoyment maybe the extra work would be appreciated.
If you are into this for the hobby aspect you might find the extra circuitry as a challenge to your skills and include it. It is all fun so do what you want since you can. The nice thing is however you build your Leach you will have a very nice amplifier, which you constructed, for the rest of your life. I sold off some extremely nice audiophile components after building my Leach and passive pre and have not regretted it once.
Tad
Bigpanda,
When I had first assembled the Leach 10 transistor clone I biased it up to 1 amp per rail. The heatsink temperature never became an issue. I used sinks with 2.5 in fin height and they handled the heat well. I have since dropped the bias, as you said, back to approximately .750 amps per rail. I use the Leach method for bias measurement outlined on his website. He reads the total rail load at the fuse with a DMM. Though this is not the preferred method here at DIY audio it works for me.
Somewhere in the group buy thread I have a picture posted of the completed amp which should give you some indication of my heatsink size. Something like 15 x 9 x 2.5 with a 3/8 inch base. Any sink that the Jens boards will fit on is going to be quite large.
Merry Christmas to all Tad
When I had first assembled the Leach 10 transistor clone I biased it up to 1 amp per rail. The heatsink temperature never became an issue. I used sinks with 2.5 in fin height and they handled the heat well. I have since dropped the bias, as you said, back to approximately .750 amps per rail. I use the Leach method for bias measurement outlined on his website. He reads the total rail load at the fuse with a DMM. Though this is not the preferred method here at DIY audio it works for me.
Somewhere in the group buy thread I have a picture posted of the completed amp which should give you some indication of my heatsink size. Something like 15 x 9 x 2.5 with a 3/8 inch base. Any sink that the Jens boards will fit on is going to be quite large.
Merry Christmas to all Tad
Greetings to all,
I would like to build leach amplifier using jens designed pcb
however I can;t open the gerber file.
I've tried gerbview and Gerb magic but it still won't open
Please advise me how to read Jen's gerber file proeprly
Thank you
I would like to build leach amplifier using jens designed pcb
however I can;t open the gerber file.
I've tried gerbview and Gerb magic but it still won't open
Please advise me how to read Jen's gerber file proeprly
Thank you
Hi,
the Leach is biased by the Optimal ClassAB method.
The device bias is determined by the Re value and the Vre across that emitter resistor.
The Total output bias for Xpairs of output devices is Vre/Re * X
Using Leach's Re=0.33 and Vre= 20mV. Then a 2pair Lo Tim has a total bias of 0.020/0.33 * 2 = 120mA.
I one keeps the same Re value and chooses to use 3pair then the total bias turns out to be ~180mA.
Going back to the standard 2pair Leach with a voltage amplifier stage, Pre-drivers and Drivers, the total quiescent current is ~150mA
the Leach is biased by the Optimal ClassAB method.
The device bias is determined by the Re value and the Vre across that emitter resistor.
The Total output bias for Xpairs of output devices is Vre/Re * X
Using Leach's Re=0.33 and Vre= 20mV. Then a 2pair Lo Tim has a total bias of 0.020/0.33 * 2 = 120mA.
I one keeps the same Re value and chooses to use 3pair then the total bias turns out to be ~180mA.
Going back to the standard 2pair Leach with a voltage amplifier stage, Pre-drivers and Drivers, the total quiescent current is ~150mA
Last edited:
yes, my consideration is long term reliability as i was not so confident about the heatsink used.....actually i was talking about my super leach amp. about the same bias as tad's.....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.