![]() |
| | #1 |
| Junior Member Join Date: Jul 2007 Location: Austria
Posts: 16
| My Amp Project: 100W 2 Channel Amp
Hey, at first sorry for my english, I'm from Austria and so... I want to introduce my actual project: A 100W 2 Channel Amp. I build the amp in context of a school project. I designed and calculated the amp by myself, but as you can see, it is very similar to a recti. First channel is Clean, second the Lead Channel. As bonus, I try to increase the second order harmonics in the PI stage, so that I get a warmer sound. Of course switchable. Now I finish the schematics, the board layouts and buy the last parts. I think, that I'll start to build the amp around September/October. If anybody has questions, please be so nice and tell me. Preamp: http://www.dobernigamps.kilu3.de/tem...nie_Preamp.pdf Board (with Loop and PI): http://www.dobernigamps.kilu3.de/temp/Board_Melli.pdf Regards Stefan EDIT: Sry, I forgot to paste the Link of the other Thread in a german forum. http://www.tube-town.de/ttforum/index.php?topic=4873.0 Last edited by Kramusha; 07-24-2007 at 05:10 PM. |
| | |
| ...and now, a word from our sponsor: |
| | #2 |
| Senior Member Join Date: Jul 2007 Location: Indianapolis
Posts: 676
|
Using headers and jumpers to select various component values is a nice trick, but I see 3-pin parts with components connected to the end pins - the middle pin is open. If your plan is to use computer-style shorting jumpers, they will only connect adjacent pins - this will leave your selection unconnected. Looking at the layout, it seems you have some other plan. You could move C1 closer to J13 to eliminate that layer change next to PAD5. I would also suggest running the trace from V1-pin6 up the middle of the tube socket, under the C1 you just moved to reach J13. That would let you eliminate the layer change between V1-8 and R18. The run from C10 to R22 could take a similar path, avoiding that layer change under R11/C5. R49 could move 'North' a bit to avoid the layer change leaving V3-pin6. These are MINOR changes that probably won't help the tone of your circuit, but they do get you to a 1-layer PCB (except for the board-to-board links, but you can bring those to the same layer, too), which is usually substantially cheaper - in fact you can etch them at home. Hope this helps! Last edited by Don Moose; 07-24-2007 at 08:21 PM. Reason: clarity |
| | |
| | #3 |
| Junior Member Join Date: Jul 2007 Location: Austria
Posts: 16
|
Thank you very much! The red lines I wanted to make as cabels. Your tips are very interessting, I will combine them with my board. The board is a 1-layer PCB, I'll etch tham at my school. That are things, I have to learn.. Stefan |
| | |
| | #4 |
| Junior Member Join Date: Jul 2007 Location: Austria
Posts: 16
|
So, the work has begun. Now I'm working on the prototype chassis (drilling and bending). The PCBs are etched, drilled and populated(? The parts are on it..) I think, the amp makes its first sounds about Christmas. I'm very exited You can see some pictures here. Stefan |
| | |
| | #5 |
| Senior Member Join Date: Jul 2007 Location: Indianapolis
Posts: 676
|
Very nice, clean work. Well done. I am a little worried about all that copper around the traces - with the very high impedances valve circuits present, it could cause you trouble in at least one of two ways. 1 - if you connect those large fields of copper to Earth, the capacitive coupling can bleed highs out of your circuit. Connecting those fields to some other circuit node would only drain the highs into that node. 2 - if you let those fields 'float', the same capacitive coupling could couple sections of the circuit such that you produce a powerful oscillator (possibly at RF frequencies). However - I am only worrying about a possibility. It is entirely possible your circuit will work brilliantly as executed - that looks like at least 2mm space between circuit traces and 'free' copper. Hope this helps! |
| | |
| | #6 | |
| Member Join Date: Aug 2006 Location: Finland
Posts: 26
| Quote:
Really nice work you've done! - Leka | |
| | |
| | #7 |
| Junior Member Join Date: Jul 2007 Location: Austria
Posts: 16
|
Yes, you are right, 2.5mm. The copper is connected to ground. I think, this is definitively the last amp with pcb. After this one, I want do one with eyelets.. I will tell you what's happening then. Stefan |
| | |
| | #8 |
| Junior Member Join Date: Jul 2007 Location: Austria
Posts: 16
|
I'm sorry, the provider has changed the urls. There are no longer schems available, but I try to upload them very soon. Stefan |
| | |
| | #9 |
| Junior Member Join Date: Jul 2007 Location: Austria
Posts: 16
|
I got finally time to update and upload the schems + board layouts. You can find them here: http://www.dobernigamps.kilu.de/projekte/melanie/ Stefan |
| | |
| | #10 | |
| Member Join Date: Aug 2006 Location: Finland
Posts: 26
| Quote:
Looks great! You've done some serious designing there. Why didn't you also make switchable bypass capacitors? Those 22n are quite big to my taste... - Leka | |
| | |
| | #11 |
| Junior Member Join Date: Jul 2007 Location: Austria
Posts: 16
|
I worked one year on this amp, one year of learning and designing.. But that's far not enough. (This means, that I haven't thought about these things enough.) I will unsolder the Coupling Caps today and put there some lugs/eyelets (Ösen...) So I can change the caps within minutes. Stefan |
| | |
| | #12 |
| Senior Member Join Date: Jul 2007 Location: Indianapolis
Posts: 676
| |
| | |
![]() |
| Bookmarks |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | |
| Display Modes | |
| |
Similar Threads | ||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| ceriatone 5e3 project | fletchyalereverb | 5 E 3 | 3 | 09-18-2007 12:40 PM |
| 100W Marshall plexi style chassis w/ plates | Marc | Flea Market | 1 | 06-06-2007 12:26 AM |
| Change of Tubes for Marshall DSL 2000 100W | Guimby | Tubes (Valves) | 0 | 03-26-2007 05:57 PM |
| 20W SLO project | chunkitup | Theory & Design | 4 | 02-28-2007 02:43 PM |
| amp project | Hennie | Build Your Amp | 7 | 09-04-2006 05:10 PM |