diyAudio Forums Archive > Top > Amplifiers > Solid State Pages: [1] 2 3  Leach clone, pretty good looking - Click HERE for Original Thread peranders http://www.delta-audio.com/Leach-Clone.htm Anyone who have built? A real nice looking pcb from Delta-Audio and Jens Rasmussen. Jens :up: :nod: JensRasmussen Hi Per, Thanks for the nice words :) I have plans for a PCB order soon, but right now my financial situation doesn’t allow it (Det er jo jul). I have uploaded the gerber data for the small version also \Jens peranders How come? As I understand you earn LOT'S of money in selling pcb's, like 1000 USD a day.... no? Sorry I must have got it wrong. :D I think the Leach design deserves a "professional" high quality pcb since the basic besign is quite good. JensRasmussen You got it wrong \Jens peranders I'll guess we will check out for a "new" Leach amp after Christmas then. keypunch Hi, If a group buy for PCBs happens for this 4 or 10 driver version I like to know. Regards, John L. Males Willowdale, Ontario Canada 23 December 2004 04:22 jacco vermeulen Same here for the 10 device boards. I love the royal number of spaces for foil caps in the powersupply section, i have loads and loads of Roederstein/Siemens MKP,MKC and MKT's. It is really great what Jan and you are doing for the people here, Jens. My brotherinlaw is becoming CEO of ISS in Copenhagen in May 2005, maybe on visiting him i can buy you guys a beer. JensRasmussen Merry Christmas, Some of my work colleagues and I plan to order some boards during January. I’ll report back when everything has been tested. \Jens Damon Hill What are the dimensions of the boards? JensRasmussen Hello The 10 transistor version is 10.7 x 25 [cm] witch is 4.2 x 10 [inches] \Jens still4given Please keep me in mind for the 10 transistor boards. Thanks, Terry still4given quote: Originally posted by JensRasmussen Hello The 10 transistor version is 10.7 x 25 [cm] witch is 4.2 x 10 [inches] \Jens Is that for a single channel board? For stereo, would I need two? Thanks, Terry JensRasmussen Hello, This is one channel per PCB, for stereo you will need 2 boards. \Jens JensRasmussen Boards are ordered... only 30 more days to go ;) \Jens Vesa Am I blind or.. Where I can find this new pcb of Leach? I think I've seen pictures of it somewhere. :scratch1: -Vesa- kilowattski Read the first post of the thread.:cannotbe: Vesa quote: Originally posted by kilowattski Read the first post of the thread.:cannotbe: I did, once more. There's shematic and gerbers. Layout seems missing :apathic: Ipanema Hi Jens, Can I replace MJL1328 and 13202 with MJL21193 and 21194? Thanks. JensRasmussen If you have these transistors available, I don’t see any reason for not using them. I plan to use either MJL3281/1302 or their newer bigger replacements MJL4281/4302. \Jens Vesa Here (see attachment) is the reason, why I asked about "missing" layout. But yes. There is picture after all. Right click -> wiew image. I dont' know if some "wiewer" missing of my PC, or why it's not wisible. Sorry about my nonsense :( Let's build some amps :smash: -Vesa- MrTransistorm quote: Originally posted by Vesa Here (see attachment) is the reason, why I asked about "missing" layout. But yes. There is picture after all. Right click -> wiew image. I dont' know if some "wiewer" missing of my PC, or why it's not wisible. Sorry about my nonsense :( Let's build some amps :smash: -Vesa- Hmmmm..... That's odd. Both images are just run-of-the-mill GIF images, but they are quite large. If you have a dialup connection it may take a while to download. Do you not have Mozilla set to automatically display images? Go to Edit > Preferences > Privacy & Security > Images. Select "Accept all images" under Image Acceptance Policy. Also click on "Manage Image Permissions" to make sure that you aren't blocking any sites unintentionally. Vesa MrTransistorm: Hi and thank you very much ;) Now that works I have now located that bug. I don't know, how Mozilla did that by itself :mad: peranders You have probably made it by mistake like I did. Do you ever right-click? Maybe you have missed once and hit "block images from this server"? Vesa quote: Originally posted by peranders Do you ever right-click? Maybe you have missed once and hit "block images from this server"? Now I remember. Probable that's just what were happened. still4given quote: Originally posted by JensRasmussen If you have these transistors available, I don’t see any reason for not using them. I plan to use either MJL3281/1302 or their newer bigger replacements MJL4281/4302. \Jens Hi Jens, Do you think the MJL4281 is superior to the MJL 3281? Do you have an idea of when the PCB's will be ready? Thanks, Terry JensRasmussen Hello Terry, I don’t really think there is any difference in performance in the voltage range I plan to use the transistors. I hope to be driving my amps at +- 68V or around that depending on what transformers I can get my hands on. The boards will arrive in mid February and then I’ll need some time to do testing. I'm really busy at work at the moment, so bare with me if I’ll be slow with the updates. I will post pictures and other information when I get the time. \Jens still4given I was planning on using my 62-0-62 toroid for this. That would be aprox 87V rails. Will that work with your design? This is the main reason I was looking at the Leach Superamp. Someone told me it could run at those voltages. Thanks, Terry jleaman I'm wondering on any one's thoughts are on the leach amp the 4 transister one or the 10 are they good for sub amps i only need a few watts like 100 or so decent ? still4given bttt JensRasmussen Hello Terry, I have simulated the circuit using 75 V rails. However I have not tested it in real life. I think that the Leach super amp is a result of output devises at the time of construction seldom could be used at more than perhaps 120 – 160V thus the need for an output ladder. I like the superamp too, and might make a layout for that sometime.... ;) \Jens acenovelty Yo jleaman and still4given, My Leach amp has been driving a pair of 12" active crossover subs for more than 10 years. Only modification was to increase the bias to 125 for a bit more low end. My Leach super amp runs quite happily on 93V rails producing 300W. But can honestly say that listening to it in a 900 square foot room at 62 Watts will make your ears ring for hours. Caps that will handle this voltage are not cheap. :$: Prosit :drink: still4given quote: I think that the Leach super amp is a result of output devises at the time of construction seldom could be used at more than perhaps 120 – 160V thus the need for an output ladder. I like the superamp too, and might make a layout for that sometime.... Huh, for some reason, I was under the impression that your board is a super amp design. Rats!:bawling: Sorry for all the trouble. I should have paid better attention.:xeye: Do you know of anyone making boards for the Superamp? Thanks, Terry still4given quote: Originally posted by acenovelty Yo jleaman and still4given, My Leach amp has been driving a pair of 12" active crossover subs for more than 10 years. Only modification was to increase the bias to 125 for a bit more low end. My Leach super amp runs quite happily on 93V rails producing 300W. But can honestly say that listening to it in a 900 square foot room at 62 Watts will make your ears ring for hours. Caps that will handle this voltage are not cheap. :$: Prosit :drink: Hi Prosit, I guess we were typing at the same time. Did you make your own boards for the superamp or do you know someone who is now making them? I have no experience making PCB's and I would hate to go to the expense of building this amp and have it fail because I messed up the boards. Any help will be greatly appreciated. Blessings, Terry acenovelty Mornin Terry, Made my own boards using Press-n-Peel Blue Transfer Film from: http://www.techniks.com/ Short tutorial: http://www.techniks.com/how_to.htm You would also need: laser printer/copier, copper clad printed circuit board material, drill bits, drill press, ferric chloride and a good measure of patience. Never found anyone producing drawings of any kind or PCB's for sale. My old drawings are simple bitmap images. If you want to go to the effort to learn this stuff, I'll dig them out. Prosit :drink: still4given Thanks Prosit, I guess I had better learn how to make PCB's if I'm going to be able to have much success in the DIY field. No time like the present. I will look into the techniks stuff. Did you use the PCB layout from the Leach Double Barrel website? http://users.ece.gatech.edu/~mleach/superamp/ I have printed that out. Looks pretty simple. Thanks again for the offer to help. Blessings, Terry acenovelty Hi Terry, Used it as a starting point. That layout works just fine, but did not suit my chassis for wiring the power transistors easily. And I used many caps/resistors that were better and not the same physical size as in the layout. So, many changes from what you are seeing. First time out, it might be better to follow Prof. Leach's board. There are many common elements to both versions. Might be a good idea to download everything and print it out for reference. When I built the first Superamp board following the destruction manual exactly, it worked the first time. Press-n-peel is so good that you should have to sacrifice only one set to get it right the second time. Much better than UV light on sensitized boards from film tracings. Prosit :drink: still4given quote: Originally posted by acenovelty Hi Terry, Used it as a starting point. That layout works just fine, but did not suit my chassis for wiring the power transistors easily. And I used many caps/resistors that were better and not the same physical size as in the layout. So, many changes from what you are seeing. First time out, it might be better to follow Prof. Leach's board. There are many common elements to both versions. Might be a good idea to download everything and print it out for reference. When I built the first Superamp board following the destruction manual exactly, it worked the first time. Press-n-peel is so good that you should have to sacrifice only one set to get it right the second time. Much better than UV light on sensitized boards from film tracings. Prosit :drink: Hi Prosit, When you say "There are many common elements to both versions", what two versions are you referring to? Is there a PCB layout for the second one you refer to? IF there is an upgrade to the original, I would rather build that one. On the press-n-peel, did you just use an iron or did you buy the press? IF I'm going to do this, I want to do it right. Thanks again, Terry jacco vermeulen The press-and-peel stuff i tried some 15y ago, the first one introduced here was called TEC2000 (not exactly sure about the 2000, blame my age :clown: ) I blew more than a package of the stuff before getting one fairly decent pcb. Even then the lines were either too thin, or too wide. Maybe the foil has been dramatically improved in a decade and a half, i got me a UV-suitcase and an etching device and that works just fine. Should the p-and-p thing not be to your satisfactory, i can make you a couple of boards if you have a decent layout, Terry. still4given Hi Jacco, Would the layout on this page be good enough to make a pcb from? I would need two. How much would you charge? Thanks, Terry WorkingAtHome Hi Terry- I have used the photo-sensitive method many times. It is quite easy. Though there are many kits that make it even easier, all you really need is an inkjet printer, some inkjet compatible overhead film (Staples), a piece of glass and a florecent light. Print out the foil pattern on the film. Use a Sharpy-type black pen to be sure it is clean (cover any bare-looking spots). Lay the film on the photo-sens boards, lay the glass on to hold it down, and expose it with the flor light (boards will come with recommened times). Boards and developer are available from many places (Digi-Key, PartsExpress etc). I think the developer is about $2US per bottle, and mixes 1-10 with water. after developing, you will see the image of your traces on the pc board. Again use the sharpy to be sure everything you want to stay on the board is there. Then etch away. Tupperware makes a good etching tank, and you can seal it up for later use. Etching works faster when heated - I use a flood lamp help over te tupperware while I gently rock it. Use ventalation or a resperator. You can do it with for small board, but I am not recommending it officially. I also find it works best to have a beer open for sipping while you etch. It can take a little while, but you must stay on top of it. Rinse with water and you are done. There are some tinning solution products out there, but I have never tried them and have read some bad things about them. I tin by hand with a soldering iron. If there are any gaps in your traces, there are a multitude of ways to fix them, but I have never had to do so with the photo boards. -b dherr555 Hello all, I want to build the leach amp.I followed a link of his sight to a PCB maker.I am gonna draw up a pcb and order some.The prices dont look to bad .Just the the first one.When I am done I will post prices and get some extras if anyone is interested.Really fast turn around too.Not sure how long it will take to draw out wiring.I was gonna copy his layout .Looks like it will around 20.00 per channel after the first ones. its a pain trying to find some of the older parts.Anyone find a good supplier for the xformer yet. still4given When you say you are going to build the Leach amp, which one are you refering to? Superamp? If so, I might be interested in the PCB's. Keep us posted please. Blessings, Terry still4given quote: Originally posted by WorkingAtHome Hi Terry- I have used the photo-sensitive method many times. It is quite easy. Though there are many kits that make it even easier, all you really need is an inkjet printer, some inkjet compatible overhead film (Staples), a piece of glass and a florecent light. Print out the foil pattern on the film. Use a Sharpy-type black pen to be sure it is clean (cover any bare-looking spots). Lay the film on the photo-sens boards, lay the glass on to hold it down, and expose it with the flor light (boards will come with recommened times). Boards and developer are available from many places (Digi-Key, PartsExpress etc). I think the developer is about $2US per bottle, and mixes 1-10 with water. after developing, you will see the image of your traces on the pc board. Again use the sharpy to be sure everything you want to stay on the board is there. Then etch away. Tupperware makes a good etching tank, and you can seal it up for later use. Etching works faster when heated - I use a flood lamp help over te tupperware while I gently rock it. Use ventalation or a resperator. You can do it with for small board, but I am not recommending it officially. I also find it works best to have a beer open for sipping while you etch. It can take a little while, but you must stay on top of it. Rinse with water and you are done. There are some tinning solution products out there, but I have never tried them and have read some bad things about them. I tin by hand with a soldering iron. If there are any gaps in your traces, there are a multitude of ways to fix them, but I have never had to do so with the photo boards. -b Hi Workingathome, So you think the photo- sensitive boards are easier than the p-n-p method? You mention kits. Link? Thanks, Terry acenovelty Hi Terry, Seems you started a bunch of questions and comments. So here goes: There are two versions of the leach amp Published by Prof Leach. The Low TIM http://users.ece.gatech.edu/~mleach/lowtim/ and the Superamp or Double Barrelled Amplifier http://users.ece.gatech.edu/~mleach/superamp/ Yes "the layout on this page" is "good enough to make a pcb", I modified it to serve my case layout, parts choice. There must be many variations of this layout given the number of students who have to build it to pass his course material. The two versions are very similar in layout and parts used. Therefore the destruction manual applies to both versions with modifications. The Press_n_Peel Blue is not the old product that many of us had problems with in the past. Having used all the other mentioned systems and being so old :geezer: that I actually hand coated the emulsion onto the copper, made reverse films, built my own UV exposure box w/timer and then developed the board in years past, it is my opinion that the current product by http://www.techniks.com/ is the most simple, least expensive in time and money and forgiving method. Traces as small as 1/32" work just fine on .5 mil copper bonded to .5mil Kapton film. That is not to say some care and diligence is not required. Refer to the FAQ's http://www.techniks.com/pnp_faq.htm You don't learn anything from doing it right the first time. Did anyone mention this can be an expensive hobby? And yes I used an ordinary iron. Tho it does have a large platen. You can make a larger platen for any iron by attaching a larger sheet of 1/8" copper sheet to the base. Now you have a square or rectangel shape to your choice. Just let the entire base reach an even temperature. These guys make transformers specifically for the leach amp. http://www.victoriamagnetics.com/ taken from the Leach web site. None of the parts are particularly older/hard to find and many can be had as free samples from http://www.onsemi.com/ Most everything else available from http://www.digikey.com/ or the locations listed on the Leach web site under the heading: Parts List and List of Suppliers. RIF The most difficult part of building the Leach amp is setting the bias current.:cheerful: Some modest observations about the Superamp on speakers that I built to take the power. You can't play it much louder than half its ability without hurting yourself. And no distortion. It will drive a pair of subwoofers beyond your needs. It will drive most any speaker you attach it to without problems, except for the temptation to keep turning it up and causing ringing of the ears. So build it with confidence and enjoy. Prosit :drink: WorkingAtHome Hi- I have not tried the Blue film. Just mentioning my experience with the photo-etching process. Since I am hearing a lot of good things about it, I will have to give it a shot. Non-photo PCB's are WAY cheaper, so it may be a better method, though I don;t have a laser printer (could just do it at work). It actually reminded me of an old toner-based system we used for a little while to make offset lithography plates, and made me cringe. That was a LONG time ago though, and has no conection with rational decision making. Hmmm. All this talk, and I just ordered photo boards for my Leach boards (again custom designed for my chassis) the other day. How does that keep happening... ;-) -b WorkingAtHome Check out the prototypeing section of the Digikey online catalog, in particular, pages 1319-1321. http://dkc3.digikey.com/PDF/T051/SectJ.pdf dherr555 I Have not decided on the low tim or super amp yet .Before I do a pcb I am trying to find some large caps and xformer. The pcb they have is a little small and doesnt leave much room for heat sinks on transistors.They say they arent needed but I would prefer they had some just to be sure I dont have any thermal problems.Having to redo pcb would be costly .Any Ideas on layout ?I am open to suggestions because it would help lower my costs if I got some extras that other people would want. I want to build a 3 channel.I cant decide if I should go with 1 channel per board or have a larger one with all.I have also seen an amp done with the outputs on the pcb with sinks instead of the large vertical ones.They said they had no problem with heating. jacco vermeulen quote: Originally posted by still4given Would the layout on this page be good enough to make a pcb from? Its up to you. In my view it is substandard, had it not been symmetrical i would take it for a layout made with rub-on tape. The output devices are not on the pcb, meaning that they need to be wired from the board. Many regard this as a risk: it may create humming or even turn the amplifier in oscillation. My fee is 1 package of 20 NO FILTER :clown: genuine American Camel cigarettes. dherr555 Thats what I was talking about .I was thinking of doing a pcb with the outputs on the board.This will make a much larger board with the cost going way up.I am on the road for 4 months so I cant make my own.It would be a pain for me to try and drill without a press(which would be no fun too).Shoulds I try a board with 2 channels on one or 1 channel each. I want to use screw type caps for filters or would large board ones work okay.Would it be good to put rectifier and those caps on there own boards.I am going to use 4 caps per channel.I want to put three channels in a box but this may be to large of an enclosure.Any suggjestions ? dherr555 Forgot to say thank you for links and info.Its gonna take me a week or so to decide on an enlcosure size and number of channels per enclosure.I want to make enough to run a 6.1 or 7.1 setup. Would be easier and cheaper to just buy a 7.1 amp but wheres the fun in that. acenovelty OK folks, "The output devices are not on the pcb, meaning that they need to be wired from the board." This is why I redesigned the layout for my case. "Many regard this as a risk: it may create humming or even turn the amplifier in oscillation." Prof. Leach cautions about this issue in the destructions. Never experienced the problem myself. Only built a half dozen or so of his amps, tho. Try this guy's solution. Website: http://home.swipnet.se/malman/LeachAmp.html Picture: http://home.swipnet.se/malman/lowtim_circuitboard.jpg dherr555 "The pcb they have is a little small and doesnt leave much room for heat sinks on transistors." Proper sized heatsinks are readily available for this purpose. One stereo amp fits nicely in a 6"Hx17"Wx15"D box. BTW, if you put the recommended resistors across the caps, there are NO turn on or turn off noises. Also in the doc's, you may learn that if a transistor dies in this amp, you do not destroy your valuable speakers. :cheerful: still4given Hi Ace, You say you built a half dozen? Were those with his design or yours? I already have a few heatsinks that I hope to use for this amp. Did the heatsinks run hot on the amps you built? I bought large heatsinks for the ESP P101 amp I built only to find that it barely gets warm, even at ear-bleed levels. It the construction details he says you need seperate heatsinks for the PNP & NPN output Transistors. Do you know what the reason for that is? That makes the case a little harder to build. Not a big deal if it's important. I just don't understand the logic. Thanks Terry PS, Did you say you might have the layout for your design? Sure like to take a look at that. :D Blessings Damon Hill I've built several Leach amps over the years with outputs wired some six inches away from the board, with no obvious stability problems or hum. Despite the necessary rat's nest of wiring, the design seems very stable. I do recommend plenty of heat sink area and good ventilation; thermal runaway was never a problem for me, but I've got enough bias on that the heatsinks get fairly warm at idle. Sustained high power output into my inefficient AR-11s gets pretty toasty. I've recently put together a channel using one of Brian Bell's boards with MJL21193/4 plastic case transistors soldered directly to the board. It eliminates a lot of wiring, looks better, and possibly benefits by shortening the path of the feedback loop--one of the claimed benefits of the GainClones. I dunno. I haven't had an opportunity to listen to this version of the amplifier and I may have wired in a mistake in the protection circuit. I see where some layouts have been done directly wiring TO-3 devices to the boards, and that should be good, too. My Ultimate Leach Amplifier may use Brian's or Jen's board in a new chassis and using premium parts throughout. But my current amplifier using the 4.5 version board sounds very good, so I'm not in much of a hurry to spend the money. acenovelty Hi Terry, Built the first one according to the original Low TIM as a test. After that I changed the layout for all the other amps. I did not use the heatsinks he recommends in the parts list. They are about 6''Hx15"L W 2" fins and run warm partly cause the bias is set to 125 for a bit more bass. They never run hot. "seperate heatsinks for the PNP & NPN output Transistors"? Haven't been able to find a reference to this with a cursurory glance thru the doc's. However since the parts list refers to Q18 - Q21, which are the output transistors and they are mounted on the same heatink as described in the Drilling the Heat Sinks section of "Construction Details", I hope you can find this info. I know of no one who has done it otherwise. I certainly did not mount them separate. I have a bitmap drawing of the layout and parts placement. It is on another computer and I will need to dig it out of those files. Sorry that's not possible just now, but I will get them to you soon. This is also a good solution: Prosit :drink: still4given Thanks guys, I'll have to look for that reference to the weperate heatsinks. I'm sure I read it but I may be confused as to where it refered to. I'll get back to you tomorrow. Blessings, Terry Damon Hill A commercial version of the Leach amplifer was built with the NPN and PNP transistors on separate heatsinks. It worked, but it's not recommended practice. (I worked for the company, briefly; that was Electronics One, Inc., in Atlanta, waybackwhen in the 70s I guess. I still have some documentation and artwork for that amplifier. It existed in at least two physically different versions; the latter version had the separate heatsinks.) acenovelty Hi Terry, The bitmap files are much too large to post here. I can sent them to you private if you wish. The final boards are in the amp mounted in a rack not so easy to get photos. Pic of the prototype board for the Superamp. I used computer power connectors to clean up the wiring rats nest. dherr555 Thank you , I guess I will not try and fix whats not broken and go with two per enclosure and use outputs on external syncs ,they can look cool on the back.i wanted to keep it in two enclosusers for 6 channels and only buy two xformers.But I think that would be a little cluttered and get toasty in one enclosure anyway.Plus I am getting old and it would probably wiegh a ton. I like those boards using computer connectors , what gauge wire did you use to go to outputs.Do you mind if I use a similiar layout for my boards ? I should have time to draw some up this week and order some.I dont enough stuff with me to etch my own.I will probably have ten made when I do because that wont be much more then the 6 I want.I will post pictures when I get them and sell them for what they cost me and shipping when I get them .I am gonna go unmasked and no screen because its almost half the price that way .Or should I spend the extra cash.Should be around 200 - 220 for 10 or like 280 withmask and screen.Not sure if they charge extra for drilling , probably do so I wont get a real idea till make them up.I dont have any software so I am gonna use expresspcb or another like them . I cant find anyone that takes gerbers that I got on the web.The couple I found want stupid money for such a small board. AndrewT Hi, cut your board costs by putting 2 channels on the same board & then sawing them apart after delivery. How about including another section for your DC block and soft start & then there's the PSU. Just try & get the join lines straight to suit the sawing. Regards Andrew T. still4given OK, I found the reference. It's under "Circuit Description" and it states "Because there are eight output transistors, two main heat sinks per channel are required. Q18, Q20, Q28, and Q30 should be mounted on one and Q19, Q21, Q29, and Q31 on the other. " I just didn't know if there was an electronic reason for this or just a physical reason. Hi Ace, I would love to see those bitmaps if you don't mind. still4given@yahoo.com Did you make a drawing for the foil traces? How did you make your boards? That's a cool idea using the computer connectors. Really make for a clean board. Hi Damon, Thanks for the info on the heatsinks. I was hoping to just use two large heatsinks for my amp. I didn't want the case to get too large. The ESP P101 I built, ended up larger than I wanted. It will be alright once mounted in a rack, but it doesn't stack very well with my other amps. Blessings, Terry jacco vermeulen Terry, for someone claiming to be a novice you sure build a nice amplifier. Looks to me like a flawless piece of work, you have an eye for detail. Secondly, for the output of your amplifier i'd think from the size of the heatsinks cranking up the bias would be a nice option. Love the skid plating, i vision your amplifier sitting on the bed of a pickup truck. Howcome i see 2 amplifier boards and 3 sets of connecting posts? acenovelty Terry, Take a look at: "Diagram showing wiring of R61 and R63 to the sockets of Q18, Q20, Q28, and Q30 in the heat sink channel." pdf and it will become quite clear. Short answer, it's both electronic and physical. I'm at work now but will send the files this eve. Simple birmap drawing of the foil and parts layout. I live near "silicon valley" where there are big surplus houses selling everything. Copper clad is $.01/ square inch double sided. 5 gallons of industrial mix ferric cloride for $10 and so on. Press-n Peel blue from my laser printer. Etched them in an old ceramic clad baking dish. The output transistors have their own boards with the same power connectors. Simple end to end pluging with almost no rats nest. Sorry, no drawings of that part. dherr555 "what gauge wire did you use to go to outputs.Do you mind if I use a similiar layout for my boards ?" 16 Guage, use the layout as you please. If you want the foil pattern for the Low Tim version, send me e-mail address that can receive over 1Meg files. "Or should I spend the extra cash.Should be around 200 - 220 for 10 or like 280 withmask and screen" Hmmm, the board in my pic cost something like 42 cents to make. 'course I had to do everything including drilling and there is no mask. Pity. Perhaps there is a clever way to import my old bitmap drawing into a suitable file for commercial board makers. Prosit :drink: jacco vermeulen If you can find an old HP flatbed plotter the mask can be printed on the pcb, even in different colors. I mentioned this on another thread before, with such a plotter the layout can be directly printed on the board, the only one needs to do is fill the pens with etch resistant ink. I found one of those plotter models in a dumpster for free, even a bunch of pens and the manual were in the box. A bitmap can not be converted to a pcb file. In my view, for such a simple layout the easiest would be to print it on a transparent. Tape it on a pc screen and redraw the lines and connections with a pcb layout program, should take no more than an hour. The option of making separate pcb's for the output devices is a good idea. Separate boards offer more flexibility, the amplifier dimensions can be smaller, the wires connecting the boards can be only an inch long. Even professional brands like Classé had several models with separate boards for gain and current amplification. WorkingAtHome http://www.diyaudio.com/forums/show...6033#post556033 still4given quote: Originally posted by jacco vermeulen Terry, for someone claiming to be a novice you sure build a nice amplifier. Looks to me like a flawless piece of work, you have an eye for detail. Secondly, for the output of your amplifier i'd think from the size of the heat sinks cranking up the bias would be a nice option. Love the skid plating, i vision your amplifier sitting on the bed of a pickup truck. How come i see 2 amplifier boards and 3 sets of connecting posts? Thanks for the kind words. I just set the bias according to Rod Elliot's directions. I'm not sure how to set the bias hotter without getting into trouble. I would be happy to try it with some instruction. The reason you only see two boards is because I had trouble with one of them at first. A drip of solder inadvertently ended up between two diodes and shorted them out destroying one of the transistors. I was running it in stereo until I sorted that out. You'll notice a empty space on one of the heatsinks. That is where the third board now resides. I built it to use as the front end of a 5.1 system but since I don't yet have a 5.1 preamp, I am running it as a stereo amp and using the center channel to drive my JBL B360 sub cabinet. Works great! Hi Ace, I'm a bit south of you. I live in Victorville, CA about 100 miles N/E of Los Angeles. I have ordered the Press-n-Peel sheets. Do I need a specific type of copper clad board? I've seen them at Fry's but I'm not sure if they will work for that. Thanks, Terry acenovelty Plain ole uncoated, non sensitized copper clad. Choose the fiberglass substrate thickness according to your needs for stiffness and mounting on stand-offs. The copper value is measured in xx ounces. Again, you have to decide how much current etc. will flow on the traces. 1-2 oz. should do fine. If you tin the board either with commercial tinning solution or just by flowing solder onto the board traces makes the work easier. The problem with producing foils for the output transistors is all the different heat sink configuration possibilities. I just drew what I needed by hand and transferred it to the board. Prosit :drink: still4given Cool! I suppose that I should get the single sided boards for this type project, correct? Also, I had bought these heatsinks for this project. Says they are for mounting 10 TO-3 transistors. I think there are eight per channel with this amp. Do these look large enough for the Leach Superamp? Thanks, Terry jacco vermeulen quote: Originally posted by still4given I just set the bias according to Rod Elliot's directions. I'm not sure how to set the bias hotter without getting into trouble. I would be happy to try it with some instruction. Says they are for mounting 10 TO-3 transistors. I think there are eight per channel with this amp. Do these look large enough for the Leach Superamp? Thanks, Terry The heatsinks will be around 0.4 degrees/watt. What voltage does the transformer on the P101 have ? Does Rod recommend a specific heatsink for the different amplifier outputs ? acenovelty No point in paying for 2nd side copper that will be etched away. You should do the basic math against the size of the heatsinks: http://sound.westhost.com/heatsinks.htm has a calculator that gives a conservative idea. I seem to remember that Prof. Leach says about .67 degrees C/W for each heat sink on the Low TIM amp. Best check tho. The spacing on your heatsink is not really optimal. This style from http://apexjr.com (not an endorsement, example only) is much better for even distribution of heat and for installing the diodes between the transistors. What you already have may work OK, but if it was mine, I wouldn't use them. Prosit :drink: still4given quote: Originally posted by jacco vermeulen The heatsinks will be around 0.4 degrees/watt. What voltage does the transformer on the P101 have ? Does Rod recommend a specific heatsink for the different amplifier outputs ? This is how Rod suggests setting the bias on the P101. Once you are sure that everything is in order, return the bias trimpot to the minimum resistance setting. Connect the amplifier to the main supply using the 100 ohm safety resistors. Apply power - again, the voltage across each 100Ù safety resistor should be between 2 and 3 volts. Carefully advance the bias pot until the voltage across each 100Ù resistor is 2V greater than your first measurement - this represents 20mA quiescent current in the MOSFET output stage. I've heard of folks running higher bias to improve bass response. I would like to try that as the bass is a little on the light side. I just don't want to hurt anything. The transformer I used in my P101 is a 45-0-45 which is giving me about +-65VDC rails. Hi Ace, I just received the heatsinks. They are too small for this project.:( I have some others that are larger but not well designed for TO-3 transistors. Are there equivalent transistors in a TO-264 or something like that? I have two of these heatsinks. I also have 4 of these heatsinks. Thanks for all your help, Terry acenovelty The "Large Aluminum Heat Sinks Allen Bradley" are probably more suited to the Superamp. However you will need to cut them in half for the heighth to make sense. There are plastic transistors that you can use. Never used anything but T03's in my amps, so don't really know what the numbers are. Prosit:drink: jacco vermeulen quote: Originally posted by still4given The transformer I used in my P101 is a 45-0-45 which is giving me about +-65VDC rails. With regular AB amplifiers maximum dissipation occurs when the output has half of the rated powersupply voltage on it. It is easy to proove with a differential equation but simple reasoning will give the same result. With a lower output voltage the voltage loss is higher, but consequently the output current lower. Vice versa, if the voltage on the output is higher, the voltage loss in the output transistors is lower but the current through the transistors will be higher. Suppose the powersupply has 40 volts on it. The load is 8 Ohms. If there is no voltage on the amplifier output the voltage loss in the output devices is 40 Volts. However, with no output voltage there is no current, the dissipation in the transistors will be Zero. Suppose the output voltage is 40 Volts. The output current will be 40 divided 8 Ohms, the current will be 5 Amps. Again, because the voltage loss in the transistors is Zero volts, dissipation will be Zero. With an output of 10 volts, the loss in the output stage will be 30 volts. At 10 volts in 8 Ohm, the current will be 1.25 Amps. Dissipation will be 30 Volts times 1.25 Amps = 37.5 Watts. With 30 Volts output, voltage loss will be 10 Volts. Current will be 30 / 8 = 3.75 Amps. Again, dissipation will be 10 Volts loss times 3.75 Amps current, equals 37.5 Watts. Now suppose the output voltage is half of the rated level, 20 Volts. Current will be 20 / 8 = 2.5 Amps. Dissipation becomes : 20 Volts loss in the output devices times 2.5 Amps current. 20 * 2.5 = 50 Watts. For your amplifier voltage is 45 Volts, after rectification half of it is around 22 Volts. Suppose the amplifier will be set up for a minimum of 4 Ohm loads which it will drive with maximum power. Maximum dissipation becomes (22 * 22) / 4 = 121 Watts. With 4 output devices rated at 150 watts the thermal factor for a transistor is 125 / 150 = 0.83. Thermal resistance for the mica layer is around 0.50. Suppose maximum heating of the transistors is limited to 100 degrees C. With an ambient temperature of 25 degrees C it leaves a safety margin of 25 degrees, maximum die temperature is 150 degrees C. 1 transistor will dissipate 121 / 4 ~ 30 watts. At 100 degrees over ambient total thermal resistance may not be larger than 100 / 30 = 3.33 Which leaves for the heatsink : 3.33 - 0.83 - 0.50 = 2 . There are 4 transistors on the heatsink each claiming 1/4 th of the heatsink. Maximum thermal resistance of the heatsink will be: 2 / 4 = 0.50 C/W Suppose your heatsinks are 0.35 C/W That is 70 % of the calculated one, at 30 degrees C over ambient the remaining 70 degrees margin will still be able to take 121 watts dissipation before the die of the transistors becomes 125 Degrees C. Raising the temperature of a 0.35 C/W heatsink by 30 degrees over ambient means putting 85 watts in it constantly. 45 Volts continuous transformer Voltage is around 62 Volts peak after rectification. 85 divided 62 = 1.37 Amps. 1.37 Amps is around 7.50 Watts in Class A in a 8 Ohm load. As half of the current goes through one output side bias in the output stage can be set to 0.34 Amps per output device. To keep on the safe side you could set the bias to 5 Watts in Class A. With average sensitivity loudspeakers of 88 dB at 1 watt, 5 watts is 7 dB higher, giving you Class A performance up to 95 dB soundpressure level. 5 watts in 8 Ohm is a bias of 0.56 amps. Normal procedure is measuring voltage drop on the source resistors of the output device. If you are using 0.22 Ohm source resistors bias is raised untill you measure 0.06 Volts drop over the resistor. If you use Rod's method, replacing the fuse by a 100 Ohm resistor again and adjusting bias to 0.56 Amps will require a power resistor of minimum 32 watts. ( 0.56 Amps in 100 Ohm equals 56 Volts drop) A safe way would be to raise bias in steps, going from the bias now to 0.20 Amps (0.64 watt)- 0.40 Amps(2.56 watt) till the maximum level, measure heatsink temperature at every step. still4given Hi Ace, Thanks so much for the drawings. May I assume that the layout drawing needs to be reversed? It appears that it is the same side view as the foil. Hi jacco, Wow! That's a lot of info. Thanks for taking the time to type all of that. Scarey thing is that I was able to grasp a lot of what you wrote. Two months ago that would have been totally Greek to me. That's encouraging. :) If I may, let me ask a couple of questions about setting the bias. Do I understand you to say that to set the bias using Rod's method would require 100 ohm 32W resistors? You mention 0.22 ohm source resistors. I'm not sure I know which ones those are. If you look here, you can see the schematic for the amp I built. It is Figure 2 - High Power Version. Am I correct is assuming that the source resistors would be R12, R13, R16 & R17? Those are 0.33 ohm. Is this where I would measure the voltage drop? Would that change the voltage drop I should see? I don't mean to highjack this thread. Hopefully this information will transfer to this Leach amp as well, as that is what I plan to build next. Thanks again, Terry PS: The speakers I listen through mostly are JBL 4425, sensitivity (1W, 1m) 91dB SPL and 4412 sensitivity (1W, 1m) 90dB SPL. acenovelty Mornin Terry, Ummm, depends on what method you use to put it on the copper. With Press-n-Peel you want to see the image right reading. Prosit :drink: still4given Yes, that makes sense. Blessings, Terry jacco vermeulen quote: Originally posted by still4given Do I understand you to say that to set the bias using Rod's method would require 100 ohm 32W resistors? Am I correct is assuming that the source resistors would be R12, R13, R16 & R17? Those are 0.33 ohm. Is this where I would measure the voltage drop? Would that change the voltage drop I should see? Both correct. The source resistors are the big green ones on the picture, the numbers you mention. I think Mr Elliot mentions somewhere in his description that these resistors make sure the bias is divided equally between both output devices ( within twice the accuracy of the resistors). The lower power amplifier only has one output transistor per side and does not require such resistors. 0.22 Ohm is the most common value for source resistors. Source (or emitter) resistors raise the output impedance and thus lower the damping factor of a power amp. With a higher number of output devices and consequently higher number of source resistors, and the lower the value of these resistors, influence on the damping factor is minimised. 0.22 would be the normal choice, Mr Elliott is very conservative with a lot of parameters, with most he reasons that enough is enough. Stability is a very important element in his designs. The schematic of the amplifier does not have the values printed, i did not bother to read the color lines on the pic. At 0.33 Ohms and the same bias voltage drop over the source resistors becomes 1.5 times as high, 0.06 * 1.5 = 0.09 Volts If you put the bias at 0.56 Amps going through a 100 Ohm resistor it will dissipate over 31 watts. Voltage drop over the resistor will be 56 volts, not much left behind it. Suppose you use a 4 Ohm resistor in place of the fuses. At 0.56 amps you will read 2.24 Volts higher over the resistor than the reading for the currents for the front end stages. With 4 ohms the first reading would be something like 0.08 to 0.12 volts Dissipation in the resistor would be a little over 1.25 watts, a 2 watt resistor would do. still4given quote: Originally posted by jacco vermeulen Both correct. The source resistors are the big green ones on the picture, the numbers you mention. I think Mr Elliot mentions somewhere in his description that these resistors make sure the bias is divided equally between both output devices ( within twice the accuracy of the resistors). The lower power amplifier only has one output transistor per side and does not require such resistors. 0.22 Ohm is the most common value for source resistors. Source (or emitter) resistors raise the output impedance and thus lower the damping factor of a power amp. With a higher number of output devices and consequently higher number of source resistors, and the lower the value of these resistors, influence on the damping factor is minimised. 0.22 would be the normal choice, Mr Elliott is very conservative with a lot of parameters, with most he reasons that enough is enough. Stability is a very important element in his designs. The schematic of the amplifier does not have the values printed, i did not bother to read the color lines on the pic. At 0.33 Ohms and the same bias voltage drop over the source resistors becomes 1.5 times as high, 0.06 * 1.5 = 0.09 Volts If you put the bias at 0.56 Amps going through a 100 Ohm resistor it will dissipate over 31 watts. Voltage drop over the resistor will be 56 volts, not much left behind it. Suppose you use a 4 Ohm resistor in place of the fuses. At 0.56 amps you will read 2.24 Volts higher over the resistor than the reading for the currents for the front end stages. With 4 ohms the first reading would be something like 0.08 to 0.12 volts Dissipation in the resistor would be a little over 1.25 watts, a 2 watt resistor would do. Hi jacco, Yes, I knew that the schematic didn't show the value, that's why I listed it. The value is in the BOM and the schematic in the construction docs but they are on a secure page. I figured he probably wouldn't like me posting that schematic. You lost me a bit. Are you saying that I can't set the bias by measuring across R12, R13, R16 or R17? Are you suggesting that I use Rod's method, except with a 4ohm 2 watt resistor? Thanks again, Terry jacco vermeulen I am saying you can do both. The most accurate is measuring voltage over one of the resistors you mentioned; R12,R13,R16,R17. You can do it Mr Elliott's way but you will need a high power resistor if you go for 100 Ohms or the resistor will melt while biasing. Mr Elliott has a sense of humor. A lot of his designs do not have that many parts. He lists the active ones in his schematics, the passive are either standard values or can be read from the high resolution pictures he presents with his projects. For those that absolutely have a thing for doing it all themselves he seems to offer a challenge. dherr555 Hey Thanks alot guys.I decided i will just make my own boards here at the Hotel I will be in for 4 months.I dont really need that much stuff . My enclosure wont be anything more then a box but I can fix that when I get home.Probably get arrested cause they will think I got a meth lab set up or something.WHats the dif between a superamp or plain leach amp ? Which is better ?I am not looking for power , I want clean sound .I was a jet engine mech for 10 years so loud isnt my thing anymore.Is photo etching better or peel and print. Any one ever just draw by had and photo etch .Frye's here has light for 22.00 with stand and has mostly sensitized boards. acenovelty dherr555, The Low TIM Leach amp as described in the doc's: http://users.ece.gatech.edu/~mleach/lowtim/ will produce 120W of very clean music power. The Superamp, given enough voltage will produce 300W+ with a higher parts count. But it is mostly the same amp. In a hotel room Press-n-Peel is definately easier. Prosit :drink: still4given quote: Originally posted by jacco vermeulen I am saying you can do both. The most accurate is measuring voltage over one of the resistors you mentioned; R12,R13,R16,R17. You can do it Mr Elliott's way but you will need a high power resistor if you go for 100 Ohms or the resistor will melt while biasing. Mr Elliott has a sense of humor. A lot of his designs do not have that many parts. He lists the active ones in his schematics, the passive are either standard values or can be read from the high resolution pictures he presents with his projects. For those that absolutely have a thing for doing it all themselves he seems to offer a challenge. Hi jacco, OK, I think I follow. With Rod's method, You have to reduce the trim pot to minimum and take a reading and then advance the trim pot until you read 2V more than the initial reading. With your method, do I need to back down the trim pot or do I just adjust until I read .09V? Does it matter which one of the resistors I take the measurement on? Thanks Hi Ace, Since you've built both, does the Low TIM have better fidelity than the Superamp? I guess there would be an advantage to the Low TIM since you can buy boards for that and you souldn't need as much heatsink. Any idea how the Leach amps compare to the P101 I just built? Blessings, Terry acenovelty Yo Terry, The two versions differ in the power output. Just about everything else is exactly the same, parts, etc. There has never been an "advantage to the Low TIM" over the Superamp for me, 'cause I always made my own boards. Fidelity? Hmmm, the Superamp has so much headroom for my normal listening, it's kinda hard to tell the difference. In truth I've never played either one over 63W on my meter. In a small room that's ear splitting. Cost is a difference. Bigger tranny, more heatsinks, more of the same inexpensive components, higher voltage caps and so on. But living where I do gives me access to mil spec parts and cheap heatsinks,caps for 15% or so of retail from say Digikey. Have never listened to the P101. Building up a six channel version of it for my boy in Hawaii. Going to let him have one of my Lexicon CP3 surround decoders to use with it. But so far have only assembled the case.(The most time consuming part for me) I'll find out then. But then I'm also starting the Krell KSA 50 clone amp and that takes my time also. Then there's the Infinity Emit tweeter I'm reinventing for the new line array speakers. And I work for a living. Your heatsink problem....... We have a steel/ metals scrap yard supplier near me who has 12' lengths of undrilled Al extrusions that look like the ones from http://apexjr.com/ less the black anodizing. He sells it by the pound. Cheap. Look around in the LA area for such a place. You can cut this stuff with a fiber blade on your tablesaw. Then it ends up exactly the length you want. Anodizing is not very expensive around here. Or just don't bother, probably won't make much of a difference. Prosit :drink: acenovelty Hi Terry, Finally found an old pic of the Superamp finished board. Prosit :drink: acenovelty And one showing the end/heatsinks. acenovelty Output transistors boards. Probably not of any use to you due to the transistor placement on the heatsink. dherr555 Hey ,thanks for all the Info guys.Cant wait till friday ,the only place around here for pcb stuff is like an hour away.I still need to find alot of the parts and not much around here for elctronic components , some places have stuff but nothing good for the output stages and no good caps. so I will end up mail ordering everything.Might wait on making pcbs till I have the parts so I know all the holes will match what I can find.Having a tool show this weekend so I am gonna get a cheap drill pressI wonder If my BJC-80 portable printer will print on those press and peel sheets.Well I will make a test board and find out.Only non-sensitized board fry's has is 6 by 4 inches.Maybe they have larger ones in person.Hee hee now I got to find some cheap speakers to test with . I got some 40 watters with me but they pretty cheap.Good enough to play with.Have to wait to I get home to play loud anyway. acenovelty Mornin dherr555, BJC-80 as in "Canon Color inkjet printer, max. 720x360 dpi"? Press-n-Peel is for laser printers/copiers only. Prosit :drink: WorkingAtHome The photo boards I just got only require a lightbulb and the developer ($2). Might be a good solution if no lasrer printer is available. I am eager to try the transfer stuff... -b still4given Hey Guys, I ordered some sheets of the Press-n-Peel. Their copper clad sheets are pretty expensive though. Do you guys know of an inexpensive place to get one-sided boards? I should have all of the other components by the weekend so I would like to be able to make these board and get started. Thanks again, Terry WorkingAtHome Parts Express has inexpensive 1ozo copper boards. DigiKey has decent prices and a better selection. -b still4given Thanks, I ordered them from Parts express. Blessings, Terry dherr555 Well I got everthing on the way but the xformer and 10000uf caps.I decided to go press and peel .The have a thrift store here with an 18.00 hp4 laser.Got a bunch of cheap tools at show.SO i am all set . Gonna have to have a yard sale when I leave this hotel .I am going to build a low tim amp. I got an older version of pc cad with a 400 component limit and 6 layers with all the other bells and whistles.Now I can also make lcd set ups for a htpc.I figure in 2 or three years I will have the mome theater set up done and then it will be time remodel it.Thank you guys.If anyone needs anything let me know .I think i got enough stuff coming to build like 4 amps except the heatsinks and xformers and large caps. dherr555 I have a pcad 2000 , does anyone have a .pcb file that will work with it .All the ones I can find are in formats that will not load.I can make my own but I am kinda itching to start plus I got figure out how to use the program.Which I should do anyhow.Will using the bitmaps and pdf's printing to press and peel make boards that you can use ? WorkingAtHome The PDFs on Pref Leach's site will print to the exact size and quality you need. -b acenovelty dherr555, Check out Terry's boards on thread Have transformer, will build, post #85, I think. His boards look like straight from the Prof. and he seems real happy with the result. Prosit dherr555 Hello , I cant find any mj15030 or 15031 . I got 2n6474 but forgot to get 2n6476's.Anyone know of a better sub then those.The NTE book says NTE274 and NTS292 for but the 274 is a dif case.I can order the 2n6476 but wonder if there are better choices. still4given quote: Hello , I cant find any mj15030 or 15031 On Semi has them. I just got some from them. Blessings, Terry WorkingAtHome NewarkInOne.com has them as well. They have all the trannies for the Leach. dherr555 Wow , thanks guys they even have links to have them in stock.Much cheaper from them the the other places I found plus free samples too .Well I am off to get some goggles so I can etch the boards I printed out .Cam out pretty good .That press and peel is pretty neet stuff.Last time I made boards was 20 years ago in highschool.A lot better now.My xformer should be here soon and I stil need to find a good enclosure for my final amps.The ones on digikey and the other sites they have are a little pricey for empty cases.I think i found a place to get .125 aluminum rack mount cases 19"x5.12"x12 for 35.00 a piece. Villaw I think it would be much easier for everyone, if someone like Jens, who is an engineer, were to order the boards for us for a profit (for his time of course). This way everyone would be happy since constructing the superamp pcbs would be difficult for everyone. Jens you can take orders and depending on that if it is worthwhile for you ie if a lot order them then you can proceed with the pcbs, if not then we can all forget about it. How about it?

Page generated in 0.24861693382263 seconds with 17 queries,
spending 0.01611972 doing MySQL queries and 0.23249722 doing PHP things.

Powered by: Search Engine Indexer and vBulletin
Copyright ©1999-2008 diyAudio.com