Here is the most recent schematic for the 50W amp with the changes I mentioned in the previous post.
I have also added a 20K resistor in parallel with the diode string. It was brought to my attention by ostripper that a failure (open) in that string or a broken wire going to it would over bias the amp. To date, this has not happened. Still, this is cheap insurance. The 20K resistor (R25 on the 50W schematic and R35 on the 100W schematic) limits the maximum bias to approximately 200Ma per output. The amp will get hot but not blow transistors or a fuse. I have not shown any thermal shut down in the schematics. Some form of overheating protection is a good idea in any amp project.
I have also added a 20K resistor in parallel with the diode string. It was brought to my attention by ostripper that a failure (open) in that string or a broken wire going to it would over bias the amp. To date, this has not happened. Still, this is cheap insurance. The 20K resistor (R25 on the 50W schematic and R35 on the 100W schematic) limits the maximum bias to approximately 200Ma per output. The amp will get hot but not blow transistors or a fuse. I have not shown any thermal shut down in the schematics. Some form of overheating protection is a good idea in any amp project.
jkeny said:
You could build 4 output stages to use as balanced stereo. This does more than doubles the parts count and price but the results are spectacular.
I will get to that schematic at some future point. Next I will post the 200W that I promised.
Oops Steve - C4 connection in 100W amp ! that one keeps biting you on the A*S.
Edit: Hmm - 4 output stages would be difficult to get the match right between each pair of balanced stages, I feel!
Edit: Hmm - 4 output stages would be difficult to get the match right between each pair of balanced stages, I feel!
Here is the most recient schematic for the 50W amp with the changes I mentioned in the previous post.
I have also added a 20K resistor in parallel with the diode string. It was brought to my attention by ostripper that a failure (open) in that string or a broken wire going to it would over bias the amp. To date, this has not happened. Still, this is cheap insurance. The 20K resistor (R25 on the 50W schematic and R35 on the 100W schematic) limits the maximum bias to approximately 200Ma per output. The amp will get hot but not blow transistors or a fuse. I have not shown any thermal shut down in the schematics. Some form of overheating protection is a good idea in any amp project.
I have also added a 20K resistor in parallel with the diode string. It was brought to my attention by ostripper that a failure (open) in that string or a broken wire going to it would over bias the amp. To date, this has not happened. Still, this is cheap insurance. The 20K resistor (R25 on the 50W schematic and R35 on the 100W schematic) limits the maximum bias to approximately 200Ma per output. The amp will get hot but not blow transistors or a fuse. I have not shown any thermal shut down in the schematics. Some form of overheating protection is a good idea in any amp project.
jkeny said:
I really thought I had corrected C4 at least twice, but I guess not. Here is the 50W with the new safty resistor.
Here is the 200W amp I promised. This schematic does not match the boards I am selling and wouldn't fit anyway. You will notice I have removed all the trim pots (except for bias) and added servo control for DC offset correction in both the VGA and output stages. The new components are LDRs and appear as rectangles containing an LED and a resistor.
This is not a conventional servo as it will have no effect on either the low frequency bandwidth or on the sound. The amp will still pass a perfect 1Hz square wave. It will correct for small DC inputs, but will amplify DC if the input level is high enough.
I will not explain the circuit at this time as I want to see what your take on this is. I will explain later. Have fun.
This is not a conventional servo as it will have no effect on either the low frequency bandwidth or on the sound. The amp will still pass a perfect 1Hz square wave. It will correct for small DC inputs, but will amplify DC if the input level is high enough.
I will not explain the circuit at this time as I want to see what your take on this is. I will explain later. Have fun.
Ah, Steve, another morsel from the teachers table - excellent, thank you. Let the discussions begin.
I can't analyse the circuit but I bet the slower response of LDRs is significant in the operation of this DC servo!
I'm very interested in a DC servo that doesn't affect the quality of the sonics!
Edit: Just one problem with schematic the -70V connection to the O/P stage seems incomplete!
I can't analyse the circuit but I bet the slower response of LDRs is significant in the operation of this DC servo!
I'm very interested in a DC servo that doesn't affect the quality of the sonics!
Edit: Just one problem with schematic the -70V connection to the O/P stage seems incomplete!
Hi Steve,
Ive been following thread with all interest, yes sir with great interest (but silently) and with 200w design, you dropped another bomb! I'm still trying to figure out your explanations how the circuit 50w/100w works and now my logic is flip flopping. Going back I just wish to know if the 50w or 100w is bridgeable and how? I must say that your design is another first to me.
Thank you in advance.
B.Regards,
mannycc
Ive been following thread with all interest, yes sir with great interest (but silently) and with 200w design, you dropped another bomb! I'm still trying to figure out your explanations how the circuit 50w/100w works and now my logic is flip flopping. Going back I just wish to know if the 50w or 100w is bridgeable and how? I must say that your design is another first to me.
Thank you in advance.
B.Regards,
mannycc
mannycc said:
I will address the issue of bridging a little more. Any amp with one hot and one ground for output can be bridged IF two channels are used. There may be an exception to that, but not that I can think of right now.
If you have two identical channels, you simply invert the input to one of the channels. Connect the grounds of the two channels together and place the load between the two hot outputs. If you are a big believer in absolute phase then you will use the non inverted output as the + output. This assumes non inverting amps to begin with.
The main drawback to this is the expense of buying twice as many channels. Another drawback is the need to increase the number of outputs, size of heat sinks and power supply transformer if you plan on driving loads of 4 ohm or below. This also adds to the cost.
There are benefits of course. Common mode noise and distortion are greatly reduced. Slew rate is increased, and the reason that it is most often done is the extra power you can achieve. Two 100W channels will theoretical yield 400W. In reality, you will get twice the 4 ohm rated power into an 8 ohm load.
Concerning the matching of the two channels used in a balanced amp as mentioned by jkeny, this is of less importance than matching transistors within one amp. The two output stages are in series with the load between them. The current flowing in that loop is the same regardless of actual balance between stages. If there is a difference in the voltage output of the two stages, then there will be a difference in power dissipation in the two stages, but the current will be the same. Unless the difference in heating caused by this effects performance of the two stages, there should be no other miss match. I do recommend using two channels as similar as possible for bridging.
jkeny said:
I will wait for more response from more people before I explain how this works.
I have corrected that -70V connection and made some minor resistor value tweaks. The main change is the input sensitivity has been increased to give rated power with 1VRMS input.
Output stage offset adjustment
Steve,
In post #288 you introduced R32 & R33 for the adjustment of offset in the output stage of the 100w amp. In post #370 you described how to fit these extra trimpots on the underside of the boards you supply.
How critical is it that these parts are fitted if matched transistors are used? How much offset could be expected if the trim pots are omitted? I ask this since the parts may not be very accessible if they are mounted under the board.
I am also wondering why you have chosen not to use this method for adjustment of output stage offset in the 50w amp.
Thanks
Dave
Steve,
In post #288 you introduced R32 & R33 for the adjustment of offset in the output stage of the 100w amp. In post #370 you described how to fit these extra trimpots on the underside of the boards you supply.
How critical is it that these parts are fitted if matched transistors are used? How much offset could be expected if the trim pots are omitted? I ask this since the parts may not be very accessible if they are mounted under the board.
I am also wondering why you have chosen not to use this method for adjustment of output stage offset in the 50w amp.
Thanks
Dave
i feel like a kill joy
but i think that the 200w is far too sophisticated to be published or to debugged out in a thread ....
something is telling me that a lots of bits and bytes to be resolved and further analysis is to be made before go public
been involved in the past in designs that worked in simulators but practice is a hell of a diferent thing and real life tests prooved that design was poor .....
at the particular design i was simply a victim of the designer who simply through a schematic on the internet that looked attractive and simple but had alot of problems ....
i hope this is not the case here .....
please forgive me there is nothing personal here .....that is just an opinion ....
but i think that the 200w is far too sophisticated to be published or to debugged out in a thread ....
something is telling me that a lots of bits and bytes to be resolved and further analysis is to be made before go public
been involved in the past in designs that worked in simulators but practice is a hell of a diferent thing and real life tests prooved that design was poor .....
at the particular design i was simply a victim of the designer who simply through a schematic on the internet that looked attractive and simple but had alot of problems ....
i hope this is not the case here .....
please forgive me there is nothing personal here .....that is just an opinion ....
Sakis,
I'm sure Steve will answer this but I don't believe you have read the thread carefully - all Steve's schematics shown so far have been from working circuits and I'm sure this is from that stable also.
In fact there were times when some simulations of his first schematics didn't show the real operation of those circuits but the simulation s/w proved to be at fault in all cases.
So I don't think this new 200W schematic should be ignored even if it does contain subtleties that are difficult to understand - this is DIY audio after all & a forum for learning!
Edit: Tinitus posted while I was writing & said it much more succinctly !
I'm sure Steve will answer this but I don't believe you have read the thread carefully - all Steve's schematics shown so far have been from working circuits and I'm sure this is from that stable also.
In fact there were times when some simulations of his first schematics didn't show the real operation of those circuits but the simulation s/w proved to be at fault in all cases.
So I don't think this new 200W schematic should be ignored even if it does contain subtleties that are difficult to understand - this is DIY audio after all & a forum for learning!
Edit: Tinitus posted while I was writing & said it much more succinctly !
- Status
- Not open for further replies.