I have most of the parts needed to build a pretty crude Gainclone. By 'crude' I mean that the parts are not of the greatest quality, available from either RatShack or my local parts dealer. I have a few questions before I get into parts, though...
I plan to make a GC with a preamp in the same enclosure, thus creating more space. However... I do not know what kind of preamp would be best. The chips I have at my disposal are: OPA627, OPA637, OPA2132, OPA2227, OPA2134, OPA658, AD823, AD648, BUF634, OPA655, and maybe some others. I was thinking of using OPA627/637 with a BUF634 (like my headphone amp) with a 1uF input coupling cap and gain of about 8. A 10k pot would be on this input. From there, I would use a 680 ohm resistor between the preamp output and the power amp input. Would I need another coupling cap for this chip? I was thinking about setting this gain to about 21. Sound ok?
Now, some notes on parts... Preamp chips were mentioned above. The couping caps would be metal film, and resistors would also be metal film, 1/2 and 1W types. Power amp chips will be OPA541's. Power caps would be NTE 1000uF's and then Vishay/Sprague 470uF, with 1uF metal film after them. Thinking I will bias the preamp output into class A via a cascode source with a few 2N5484's. Would it be beneficial to bias the power amp outputs into class A as well? Will this make a decent sounding amp like this?
On a heat sink note... I don't know how much heat sinking I need. My local parts store does not have heat sinks, so I could either buy those little ones from RadioShack or build my own (something I don't know how to do). What can I do? What kind of enclosure should I use?
Thanks
I plan to make a GC with a preamp in the same enclosure, thus creating more space. However... I do not know what kind of preamp would be best. The chips I have at my disposal are: OPA627, OPA637, OPA2132, OPA2227, OPA2134, OPA658, AD823, AD648, BUF634, OPA655, and maybe some others. I was thinking of using OPA627/637 with a BUF634 (like my headphone amp) with a 1uF input coupling cap and gain of about 8. A 10k pot would be on this input. From there, I would use a 680 ohm resistor between the preamp output and the power amp input. Would I need another coupling cap for this chip? I was thinking about setting this gain to about 21. Sound ok?
Now, some notes on parts... Preamp chips were mentioned above. The couping caps would be metal film, and resistors would also be metal film, 1/2 and 1W types. Power amp chips will be OPA541's. Power caps would be NTE 1000uF's and then Vishay/Sprague 470uF, with 1uF metal film after them. Thinking I will bias the preamp output into class A via a cascode source with a few 2N5484's. Would it be beneficial to bias the power amp outputs into class A as well? Will this make a decent sounding amp like this?
On a heat sink note... I don't know how much heat sinking I need. My local parts store does not have heat sinks, so I could either buy those little ones from RadioShack or build my own (something I don't know how to do). What can I do? What kind of enclosure should I use?
Thanks
trespasser_guy said:
Just do whatyever you think it's right and experiment a lot. Nobody can give you a clear answer on those questions.
Don't worry about heatsinking, a piece of aluminum or whatever you can get from RS should be fine. If you have so many chips at your disposal, try to test some of them in a circuit and find out for yourself which ones sounds good. I tend to be interested in OPA627 recently, but didn't try it yet.
one idea
I've been researching gainclones as well. One idea I just came across is from this integrated amp:
http://members.ozemail.com.au/~joeras/
It uses a tube cathode follower buffered passive preamp. The gainclones can be made to have lots of gain so an active preamp isn't necessary.
A similar passive preamp (the "TLP") is offered by AKSA for their amplifiers. The tubed passive pre- provides some warmth and dimensionality. I can't imagine it'd be terribly expensive to implement -- you could also simply buy the AKSA kit since it provides everything except the enclosure and it's cheap at that.
I've been researching gainclones as well. One idea I just came across is from this integrated amp:
http://members.ozemail.com.au/~joeras/
It uses a tube cathode follower buffered passive preamp. The gainclones can be made to have lots of gain so an active preamp isn't necessary.
A similar passive preamp (the "TLP") is offered by AKSA for their amplifiers. The tubed passive pre- provides some warmth and dimensionality. I can't imagine it'd be terribly expensive to implement -- you could also simply buy the AKSA kit since it provides everything except the enclosure and it's cheap at that.
trespasser_guy said:
You can make a VERY nice Gainclone with quality parts. However, according to my experience, "crude" parts can also make good amplifiers. So, please go on.
I would not think about the preamp with gainclone because most of the power op-amps for gainclone make good quality sound by themselves. If you still want to use a preamp, I would recommend those from Texas Instruments such as OPA134. You can get it by UPS 2nd day from their nice sample program. I was pretty much satisfied by OPA134 family op-amps. I don't understand why people don't like op-amps for audio application.
You wouldn't need a very big heatsinks for gainclone unless you want to play it VERY loud with VERY low sensitivity speakers.
I have tried couple of heatsinks with LM3875 and LM1875 and concluded that direct mounting of the chips to the enclosure was good enough for my application. Don't forget to isolate the chips from the enclosure with insulating pads if you will use non-isolated chips.Thanks all!
I found a huge heatsink on eBay for pretty cheap and was about to bid before I read this where I see a huge heat sink is not really needed. I will probably fashion my own crude heat sink with a couple pieces of aluminum from Lowe's, then mount it in a hardwood case.
My primary source would be a Sony CDP-CE375 cd-player, so would I need a preamp? I am guessing no... I don't know why, but for some reason thought a premap would be necessary.
I do have a question on power supplies that just came to me... The 1000uF's are in the amp section, and basically no power supply section exists so to speak. Well, do some people create separate power supplies using large value caps, then run that power to the rails of the amp where it hits 1000uF more by the chips? If so... I may order a couple large (10,000uF+) caps and use them in a power supply section, then use the 1000's by the chips.
Although I want to do this simply and cheaply, I may just order a few things from Digi-Key... namely caps. I have always heard good things about Panasonic FC caps, and they would probably be better than the NTE's. Any opinions on Nichicon UPW's? Unfortunately, DK does not sell orange drop film caps, which I'd love to use on inputs. Though, they do have Panasonic film caps, which are supposedly fairly decent.
I worry about the gain... doesn't having it set to 21 cause some distortion?! I do a lot of work with low power opamps and circuits using them. I designed a nice headphone amp, which I cannot afford to build to test properly. It was breadboarded and performed well... Anyway, when I set the gain of an opamp higher than, say, 15, I got a significant amount of distortion.
A question on the OPA541... how much power (roughly) can one of these put out? My power supply will be 17-0-17VAC@4.7A, so once rectified I get about 24VDC, but that times two for each rail, times 4.7A, I have a lot of headroom power-wise, correct?
I also saw that some people have gotten better sound by connecting a resistor from V- to the opamp output to effectively bias the amp's output into class A. Is the difference noteable? Would it be even better to use a cascode current source?
Thanks
I found a huge heatsink on eBay for pretty cheap and was about to bid before I read this where I see a huge heat sink is not really needed. I will probably fashion my own crude heat sink with a couple pieces of aluminum from Lowe's, then mount it in a hardwood case.
My primary source would be a Sony CDP-CE375 cd-player, so would I need a preamp? I am guessing no... I don't know why, but for some reason thought a premap would be necessary.
I do have a question on power supplies that just came to me... The 1000uF's are in the amp section, and basically no power supply section exists so to speak. Well, do some people create separate power supplies using large value caps, then run that power to the rails of the amp where it hits 1000uF more by the chips? If so... I may order a couple large (10,000uF+) caps and use them in a power supply section, then use the 1000's by the chips.
Although I want to do this simply and cheaply, I may just order a few things from Digi-Key... namely caps. I have always heard good things about Panasonic FC caps, and they would probably be better than the NTE's. Any opinions on Nichicon UPW's? Unfortunately, DK does not sell orange drop film caps, which I'd love to use on inputs. Though, they do have Panasonic film caps, which are supposedly fairly decent.
I worry about the gain... doesn't having it set to 21 cause some distortion?! I do a lot of work with low power opamps and circuits using them. I designed a nice headphone amp, which I cannot afford to build to test properly. It was breadboarded and performed well... Anyway, when I set the gain of an opamp higher than, say, 15, I got a significant amount of distortion.
A question on the OPA541... how much power (roughly) can one of these put out? My power supply will be 17-0-17VAC@4.7A, so once rectified I get about 24VDC, but that times two for each rail, times 4.7A, I have a lot of headroom power-wise, correct?
I also saw that some people have gotten better sound by connecting a resistor from V- to the opamp output to effectively bias the amp's output into class A. Is the difference noteable? Would it be even better to use a cascode current source?
Thanks
I added 10,000 to my 1.000u rails and it actually stopped shining. And cables from PS to the amp make big difference indeed. I choose the cables according to the feedback resistor type I use to achieve most neutral sound.
Here's the little graph showing different power levels with regard to supply voltage for LM3875. My preference now is about 22v AC with 400VA Plitron. It seems to be more dynamic than 18V rails gainclone. You also stated interest in using OPA541. I tried OPA549 and prefer LM3875, because it produces more emotions with the music. OPA maybe sounds more correct, but more lifeless also.
And cables from PS to the amp make big difference indeed. I choose the cables according to the feedback resistor type I use to achieve most neutral sound.Here's the little graph showing different power levels with regard to supply voltage for LM3875. My preference now is about 22v AC with 400VA Plitron. It seems to be more dynamic than 18V rails gainclone. You also stated interest in using OPA541. I tried OPA549 and prefer LM3875, because it produces more emotions with the music. OPA maybe sounds more correct, but more lifeless also.
Attachments
If you had only Parts Express and Mouser to order from... which caps and resistors would you go with? I am putting a limited, unlimited budget up here. I mean, I cannot spend $100 on this... but I figured it up with a Solen 3.3uF input cap, Vishay/Dale 1% 1/2W resistors, a cheap ($2.35) 10k Xicon audio pot from Mouser, Nichicon UPW power caps, and fairly cheap gold plated RCA and binding posts. The total was about $30.
What power caps are best from Mouser? Anything I should change parts-wise?
What power caps are best from Mouser? Anything I should change parts-wise?
Why only them? For your filter caps I would go with the Panasonic FC 1000uF/50V for $2.40 apiece from DigiKey. You can Dale NH65 resistors from them as well and get some Mills 10Kresistors from PartExpress. Input caps are up to your preference but a nice polypropylene or polycarbonate from either of the companies would work well.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.