Hello folks,
For some years now I wanted to build an high-power amp, but haven't been able to do so, due to my lack of money and knowledge. Now my education has given me the knowledge and I found some of the expencive peices of an amp.
I have 2 trafos, +/-35V, 550VA, some big caps (10,000µF) and big heatsinks.
Could anyone of you help me with an amp desing (also a bit hifi if possible), about 2x300W?
For some years now I wanted to build an high-power amp, but haven't been able to do so, due to my lack of money and knowledge. Now my education has given me the knowledge and I found some of the expencive peices of an amp.
I have 2 trafos, +/-35V, 550VA, some big caps (10,000µF) and big heatsinks.
Could anyone of you help me with an amp desing (also a bit hifi if possible), about 2x300W?
The Leach Super Amp at http://users.ece.gatech.edu/~mleach/superamp/ can give you 300 Watts, but Leach says you need 93 volt rails. This is a good design.
I don't know what you mean by your transformer rating - do you get +/- 35 VDC out of it after rectification, or is is a center-tapped transformer that gives two 35 VAC outputs, and will thus give you 70 VAC when connected in series? If this is the case, then that 70 VAC will convert to 98 volts, which is probably ok to use with the Leach Amp and will clearly get you your 300 W with this design.
I should probably recommend that you build a simpler amp first. A 300 W amp is a huge project, even for an EE (if that is what you are), and the voltages you are dealing with can be dangerous. It might be best to first build a regular Leach Amp, say, and then upgrade if you find it necessary.
I am sure others will recommend other designs - I just happen to be partial to the Leach designs.
Good luck!
I don't know what you mean by your transformer rating - do you get +/- 35 VDC out of it after rectification, or is is a center-tapped transformer that gives two 35 VAC outputs, and will thus give you 70 VAC when connected in series? If this is the case, then that 70 VAC will convert to 98 volts, which is probably ok to use with the Leach Amp and will clearly get you your 300 W with this design.
I should probably recommend that you build a simpler amp first. A 300 W amp is a huge project, even for an EE (if that is what you are), and the voltages you are dealing with can be dangerous. It might be best to first build a regular Leach Amp, say, and then upgrade if you find it necessary.
I am sure others will recommend other designs - I just happen to be partial to the Leach designs.
Good luck!
Thanks for the reply,
The transformers are indeed centertapped ones with two windings of 35VAC.
I've been looking around the internet some time now and found the M250 at p://www.astro.uu.se/~marcus/private/m250.html which has been discussed here also a while ago. Is that one of good quality?
I've made a lot of integrated amps already but am rather new to full discrete desing. Other replies are very welcome.
The transformers are indeed centertapped ones with two windings of 35VAC.
I've been looking around the internet some time now and found the M250 at p://www.astro.uu.se/~marcus/private/m250.html which has been discussed here also a while ago. Is that one of good quality?
I've made a lot of integrated amps already but am rather new to full discrete desing. Other replies are very welcome.
Hey Sander,
Is that +/- 35V DC? or is that AC volts? If that is AC volts, after rectification and filtering, you'll get around 43 Volt rails in which you can get about 160W if your transformers don't drop too much voltage at full load. I built a couple 500W amps and didn't have the full rated power supply and didn't get the rated power output either. I'm only getting about a 118 Volt PK-PK into 4 ohms and that is only about 380W or so. With those rails, you will have a great 125-180W amp unless you drive a 2 ohm load!!!!
Good luck,
Chris
Is that +/- 35V DC? or is that AC volts? If that is AC volts, after rectification and filtering, you'll get around 43 Volt rails in which you can get about 160W if your transformers don't drop too much voltage at full load. I built a couple 500W amps and didn't have the full rated power supply and didn't get the rated power output either. I'm only getting about a 118 Volt PK-PK into 4 ohms and that is only about 380W or so. With those rails, you will have a great 125-180W amp unless you drive a 2 ohm load!!!!
Good luck,
Chris
Here's a thought
If you're new to amp building, start by making a stereo amp using one of your transformers, running off +/-50Vdc rails (they're 35-0-35V AC, right?). I did just this, see
http://www.diyaudio.com/forums/showthread.php?s=&threadid=18945; or you could look for a MOSFET design as they're usually more resistant to thermal abuse than bipolars.
Once you're happy it's not going to overheat or explode, you can turn it from a 2x180W/4ohm amp into a 1x360W/8ohm amp by bridging it.
Build another identical one with your other transformer and you'll have two >300W monoblocks, no spare transformers, and hopefully no fried underpants...
Cheers
IH
If you're new to amp building, start by making a stereo amp using one of your transformers, running off +/-50Vdc rails (they're 35-0-35V AC, right?). I did just this, see
http://www.diyaudio.com/forums/showthread.php?s=&threadid=18945; or you could look for a MOSFET design as they're usually more resistant to thermal abuse than bipolars.
Once you're happy it's not going to overheat or explode, you can turn it from a 2x180W/4ohm amp into a 1x360W/8ohm amp by bridging it.
Build another identical one with your other transformer and you'll have two >300W monoblocks, no spare transformers, and hopefully no fried underpants...
Cheers
IH
The M250 is more like the regular Leach Amp, power wise anyway. The Leach Amp uses +/-56 volt rails and is rated at about 120 Watts into 8 Ohms. Marcus rates the M250 at 250 Watts into 4 Ohms. Into 8 Ohms, it would be 125 Watts. I cannot comment on the sonic differences, but I like my old Leach Amp and have spent the last three years considering upgrading it from a Version 3 (but with a Version 1 power supply, so I only get about 50 Watts) to 4.5. Someday......
With 35 VAC transformers, you can get out either about 49 or 98 VDC rails. Both amps would probably work ok at 49 volts and give on the order of 50 Watts into 8 Ohms, but only the Leach Super Amp would survive with 98 volts, and is the only one of the two that would give you your desired 300 Watts into 8 Ohms.
Again, a high powered amp is a huge project (and dangerous) and unless you have experience, I think starting with a less ambitious design would be best. You will learn all kinds of things while building, so why not learn on a cheap amp than on your dream amp? But whatever you decide, have fun.
With 35 VAC transformers, you can get out either about 49 or 98 VDC rails. Both amps would probably work ok at 49 volts and give on the order of 50 Watts into 8 Ohms, but only the Leach Super Amp would survive with 98 volts, and is the only one of the two that would give you your desired 300 Watts into 8 Ohms.
Again, a high powered amp is a huge project (and dangerous) and unless you have experience, I think starting with a less ambitious design would be best. You will learn all kinds of things while building, so why not learn on a cheap amp than on your dream amp? But whatever you decide, have fun.
Just do it 
)
Hi ,
I just finish building the M250 Amplifier ...)
GOOD DESIGN !!! VERY GOOD DESIGN !!!
I don't know what to say .. if a 15 year's old kid do it .. why you couldn't do it ?? No kidding .. i have only 15 years and i do it very good ...
Be carefull with the bias current ... !!
I have some problem's with one channel .. it have a little distorsion on the 60 - 120 HZ Band and the board with the distorsion it's getting hotter that the other ..
Look at my post on the M250 AMP .. maybe someone will help ..
)Hi ,
I just finish building the M250 Amplifier ...
)GOOD DESIGN !!! VERY GOOD DESIGN !!!
I don't know what to say .. if a 15 year's old kid do it .. why you couldn't do it ?? No kidding .. i have only 15 years and i do it very good ...
Be carefull with the bias current ... !!
I have some problem's with one channel .. it have a little distorsion on the 60 - 120 HZ Band and the board with the distorsion it's getting hotter that the other ..
Look at my post on the M250 AMP .. maybe someone will help ..
Hi ,
I measure by the ear with a tone generator ( on the computer )
)
I hope that the power supply to be the problem ... but it isn't ...
Bias ?? Maybe bias current ??
I tested all the componenet's on the board .. they are all good ...
power supply checked .. only bias ..
AND THE BOARD WITH THE PROBLEM IS GETTING VERY HOT !!!
I measure by the ear
with a tone generator ( on the computer ))I hope that the power supply to be the problem ... but it isn't ...
Bias ?? Maybe bias current ??
I tested all the componenet's on the board .. they are all good ...
power supply checked .. only bias ..
AND THE BOARD WITH THE PROBLEM IS GETTING VERY HOT !!!
If it's getting hot, when there sure is a problem
However oscillation will make your output hot until it blows up....
You have two boards; One is ok and the other is not...
Place the two boards side by side, and carefully compare component colour code/type numbers etc......
The turn the boards upside down and carefully compare the PCB's and all the solderings. I normally uses a magnifier glass for this, and it have saved me many troubles
If you haven't found any fault, then power up both prints and carefully check DC voltage ratings on each components, making notes on a schematic (both for the good and the bad working PCB). If you find any interesting deviations, let us know
However oscillation will make your output hot until it blows up....
You have two boards; One is ok and the other is not...
Place the two boards side by side, and carefully compare component colour code/type numbers etc......
The turn the boards upside down and carefully compare the PCB's and all the solderings. I normally uses a magnifier glass for this, and it have saved me many troubles
If you haven't found any fault, then power up both prints and carefully check DC voltage ratings on each components, making notes on a schematic (both for the good and the bad working PCB). If you find any interesting deviations, let us know
If you made your own boards there is a chance that a microscopic trace is left between two seperate traces. The other possiblity is if you overheated the trace during the soldering of a part you can get 'seapage' inside the layers of the board that you cant see from the surface. Measure between the E, B, and C of all transistors with the amp off to see if there is a short. The diode setting on your meter is good for this.
It will help to change your view that the two boards are exactly the same. They are not. One is getting hot, the other not. So, there IS a difference. Now find it. Jan's advice above is a good starting point. Remember: you are SURE there is a difference. Where? I noted you didn't change the output transistors because they didn't get hot. Yet, one of them could cause the extra current drain from the driver. These are important pointers.
Jan Didden
Jan Didden
Again ...
I clean the soldring , check evey little component , changing all the capacitors again ... verifing all the wireing .
I don't know what is telling me that the bias current it's the sucker ...
That board was burned because i set the bias to low resistance , my output blow up , that was in the past . I repair the board , changed all that was damaged .. only thing i didn't change was the resistors ( they all are good ) and the bias current pot .. maybe that pot is damaged .. When i mesure it with the multimeter it's working fine .. but i'am thinking when it's in use .. that pot may create problems ?
And it's normaly that you board to heat if the bias is set to maximum !?
And my distorsion .. may be a crossover one .. that's way i thinked at BIAS .
I think it should be normaly the board to heat at a high bias current .. not a low one ...
Thank's very much . Mihai .
I clean the soldring , check evey little component , changing all the capacitors again ... verifing all the wireing .
I don't know what is telling me that the bias current it's the sucker ...
That board was burned because i set the bias to low resistance , my output blow up , that was in the past . I repair the board , changed all that was damaged .. only thing i didn't change was the resistors ( they all are good ) and the bias current pot .. maybe that pot is damaged .. When i mesure it with the multimeter it's working fine .. but i'am thinking when it's in use .. that pot may create problems ?
And it's normaly that you board to heat if the bias is set to maximum !?
And my distorsion .. may be a crossover one .. that's way i thinked at BIAS .
I think it should be normaly the board to heat at a high bias current .. not a low one ...
Thank's very much . Mihai .
Mihai,
Jan is telling you some very good information and you should do as he says. I had to do the same thing on some amps that I built, but in my particular case, it didn't help with troubleshooting but I sure learned my amp design inside and out. THEN, I put a scope across the output and there it was.... It was singing like a bird. I tried putting a capacitor in different places but the bias was my problem too. In the bias circuit was a capacitor for filtering and anti-oscillation and all I did was increase its value by about 3X from 100pf to 320pf and it stopped the problem. Changing parts won't necessarily fix this problem. This is where the "black magic" aspect comes in, sometimes it just happens and you filter it and go on. If you are using PC boards, that's about all you can do but if it is your own layout, it will be in your parts placement, which can be changed without parts changes from the original design.
Jan, great ideas and posts!
Good luck,
Let us know what you do to fix the problem!
Chris
Jan is telling you some very good information and you should do as he says. I had to do the same thing on some amps that I built, but in my particular case, it didn't help with troubleshooting but I sure learned my amp design inside and out. THEN, I put a scope across the output and there it was.... It was singing like a bird. I tried putting a capacitor in different places but the bias was my problem too.
In the bias circuit was a capacitor for filtering and anti-oscillation and all I did was increase its value by about 3X from 100pf to 320pf and it stopped the problem. Changing parts won't necessarily fix this problem. This is where the "black magic" aspect comes in, sometimes it just happens and you filter it and go on. If you are using PC boards, that's about all you can do but if it is your own layout, it will be in your parts placement, which can be changed without parts changes from the original design.Jan, great ideas and posts!
Good luck,
Let us know what you do to fix the problem!
Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.