Hi Andrew,
I'm not sure what it was doing. It is not doing it now. I can run the sinewave genny up full and the wave is perfect. It could just be because of the way I have everything hooked up with all the open wiring and everything just clipped together. Also, one thing i just thought of is that the wires running from the board to the heatsink are twisted on the side that didn't clip and not on the one that did. I may try twisting those. I don't have the metal right now to finish the case so i might as well spend the time and get things as good as I can while everything is easily accessable.
Blessings, Terry
Hello again guys.
My conclusions so far are those:
The Leach Superamp requires well matched transistors and is susceptible to oscillations if not built super tidy. No protections built in ?
The Dx amp... F**k my life, i didn't understand much of it, except the fact that it's a tuned Leach Amp. Protections?
The Apex B500 appears to be the best one for my cause. Although, the simplicity makes me think it won't perform as well, sound quality wise, as others. The fact that it uses a dual opamp package is ok because the actual opamps inside the package are matched, but the sq will be very dependent on opamp quality and i am not sure i can find very good opamps around here. I also can't figure out which is the latest schematic and board design for the amp or the protection circuit. I also don't clearly understand which protections it has built in and how well they are implemented.
Am i missing anything regarding those amps?
Are there any other amps worth looking at?
Overall, which would be better?
My conclusions so far are those:
The Leach Superamp requires well matched transistors and is susceptible to oscillations if not built super tidy. No protections built in ?
The Dx amp... F**k my life, i didn't understand much of it, except the fact that it's a tuned Leach Amp. Protections?
The Apex B500 appears to be the best one for my cause. Although, the simplicity makes me think it won't perform as well, sound quality wise, as others. The fact that it uses a dual opamp package is ok because the actual opamps inside the package are matched, but the sq will be very dependent on opamp quality and i am not sure i can find very good opamps around here. I also can't figure out which is the latest schematic and board design for the amp or the protection circuit. I also don't clearly understand which protections it has built in and how well they are implemented.
Am i missing anything regarding those amps?
Are there any other amps worth looking at?
Overall, which would be better?
Hi Terry: I feel a bit better now 
.
Someone at Antek messed up labelling this one... I have a couple of Antek 65-0-65's. I better measure them, they better not be 75-0-75 and blow up any caps.
One thing about Antek is they are designed on a 50Hz supply, with us in the US using 60Hz, there is a 20% increased flux available or should i say, you can use this as a 1.2kVa Toroid, since every peak of the AC sinewave will come 20% quicker than running this on a 50Hz mains supply.
.Someone at Antek messed up labelling this one... I have a couple of Antek 65-0-65's. I better measure them, they better not be 75-0-75 and blow up any caps.
One thing about Antek is they are designed on a 50Hz supply, with us in the US using 60Hz, there is a 20% increased flux available or should i say, you can use this as a 1.2kVa Toroid, since every peak of the AC sinewave will come 20% quicker than running this on a 50Hz mains supply.
I have a few Antek toriods. When I first got into DIY I ordered a couple from him and they came wrong. Rather than sending them back he gave me really good price to keep them so I did. I still have two 25-0-25 300va sitting waiting for a project. Most of the amps I have built have toriods from John.
Brlmat,
Please don't count out the Superamp based on my experience. This was my second DIY project and I actually built it while I was building my first, which was an Elliot Sound P101. The funny thing is, I started much like you are because I bought this transformer by mistake and then began looking for an amp to build to go with it. you can see my journey here. They were my first home made boards. I made some mistakes with parts and couldn't get it working as a result. I just shelved it and didn't touch it again until now. I fully believe that if you are careful during assembly, it will not be a problem at all. The leach is a very good sounding amp. I have been driving two sets of JBL's with this for two days now and it is barely warm. I can't say how it compares to the other amps mentioned, I would just hate for you to be scared off because of the issues I've had. They were mostly caused by my inexperience and mistakes that resulted in numerous device exchanges. Not a good scenario when using home etched boards. If I had the time an money, I would probably build another one so I could get it right the first time. I have met guys who have built more than one. That says something for it.
Blessings, Terry
Brlmat,
Please don't count out the Superamp based on my experience. This was my second DIY project and I actually built it while I was building my first, which was an Elliot Sound P101. The funny thing is, I started much like you are because I bought this transformer by mistake and then began looking for an amp to build to go with it. you can see my journey here. They were my first home made boards. I made some mistakes with parts and couldn't get it working as a result. I just shelved it and didn't touch it again until now. I fully believe that if you are careful during assembly, it will not be a problem at all. The leach is a very good sounding amp. I have been driving two sets of JBL's with this for two days now and it is barely warm. I can't say how it compares to the other amps mentioned, I would just hate for you to be scared off because of the issues I've had. They were mostly caused by my inexperience and mistakes that resulted in numerous device exchanges. Not a good scenario when using home etched boards. If I had the time an money, I would probably build another one so I could get it right the first time. I have met guys who have built more than one. That says something for it.
Blessings, Terry
Hi Andrew,
I just read this again and thought I should comment. The amp wasn't clipping at 145W. I had some sort of oscillation or feed back thing happening. Only one side did it and now is isn't. My Sinewave generator only puts out 1.5VAC and that will only take the output to 39.9VAC. There is no clipping at that point. I guess I should hook up the generator though a preamp so I can drive the amp to clip. When my dummy load arrives I will try that. I don't want to do it through my speakers.
No, i won't count out the Superamp. I understand why you had those problems and it does not worry me as i am not going to have my boards etched. They are going to be cnc'd. I don't like the risk of corroding too much or too less. I'll just take a layout, draw it in autocad, which is fairly simple and send it to the cnc machine programmed to the exact dimensions. I'll only use the peak of the drill when cnc-ing it so that i can get the lines pretty close to each other if that's the case.
Also, i already talked to the guys at the electronics shop and they told me i can measure the transistors before buying so that's gonna be a big save if i am required to match them.
My questions about this amp:
1.Does it have protections? Which?
2.Output power in 8 ohms at +- 75v
3.How well should the transistors be matched?
4.Where the hell can i find the schematic and layout? (top, bottom, top and bottom)
The same questions apply to Apex B500 if anyone's willing to answer them...
Thank you!
Also, i already talked to the guys at the electronics shop and they told me i can measure the transistors before buying so that's gonna be a big save if i am required to match them.
My questions about this amp:
1.Does it have protections? Which?
2.Output power in 8 ohms at +- 75v
3.How well should the transistors be matched?
4.Where the hell can i find the schematic and layout? (top, bottom, top and bottom)
The same questions apply to Apex B500 if anyone's willing to answer them...
Thank you!
Hi Guys
Terry, correct connections of the bread board even if poorly laid out will provide full power and clipping at the same level as the properly dressed completed amp.
It is more likely that the side that did not clip so early did not have the load connected.
As an aside, from the notes in this thread combined with other discussions I've had with other builders, it does not sound like Antek is very competent. Supporting this is a rather poor PT design they have aimed at tube amp builders. That PT has no bias winding and is intended to be used with a half-bridge rectifier. The discontinuous conduction loading on each winding caused by this rectification is about the worst you can mate with a toroid.
The price of toroidal winding equipment dropped a few years ago and it seems like a lot of people have gotten into that biz who maybe shouldn't have. Learning curves are one thing, and mislabeling is human error, but incorrect designs are right out.
Note that the "Dx Blame" is a reduced performance version of Doug Self's "Blameless" amp. The VAS current source has been replaced by a bootstrap. The diff-amp current source is no longer active, rather the buffered passive type. Nothing to do with Leach at all.
The Apex SR100 is similar to the Hitachi mosfet app circuit, but with BJT outputs and the input and VAS cascoded, with the bandaid of a DC servo added. To me, the _need_ of a DC servo reflects poor circuit design.
The Apex B500 is a simplified dorm of the Yorkville Sound PA amps with the anti-clip circuit found in a few other models and brands added in. Power amps that require opamps for their input were done in the '70s by BGW and others. Most had really poor performance. The YS amps and all the others in this group are built to be tough, not to be hifi.
Have fun
Kevin O'Connor
Terry, correct connections of the bread board even if poorly laid out will provide full power and clipping at the same level as the properly dressed completed amp.
It is more likely that the side that did not clip so early did not have the load connected.
As an aside, from the notes in this thread combined with other discussions I've had with other builders, it does not sound like Antek is very competent. Supporting this is a rather poor PT design they have aimed at tube amp builders. That PT has no bias winding and is intended to be used with a half-bridge rectifier. The discontinuous conduction loading on each winding caused by this rectification is about the worst you can mate with a toroid.
The price of toroidal winding equipment dropped a few years ago and it seems like a lot of people have gotten into that biz who maybe shouldn't have. Learning curves are one thing, and mislabeling is human error, but incorrect designs are right out.
Note that the "Dx Blame" is a reduced performance version of Doug Self's "Blameless" amp. The VAS current source has been replaced by a bootstrap. The diff-amp current source is no longer active, rather the buffered passive type. Nothing to do with Leach at all.
The Apex SR100 is similar to the Hitachi mosfet app circuit, but with BJT outputs and the input and VAS cascoded, with the bandaid of a DC servo added. To me, the _need_ of a DC servo reflects poor circuit design.
The Apex B500 is a simplified dorm of the Yorkville Sound PA amps with the anti-clip circuit found in a few other models and brands added in. Power amps that require opamps for their input were done in the '70s by BGW and others. Most had really poor performance. The YS amps and all the others in this group are built to be tough, not to be hifi.
Have fun
Kevin O'Connor
Last edited:
Here is the website for the Superamp. It has all the info you are asking for. Note that most of the construction info is on the Low TIM pages which are linked to in the Superamp docs. It would be good to read through both before starting. Dr Leach was a professor and used these amps for teaching purposes so there is a lot of really good info.
So far I have not been able to drive my to clipping. As I said earlier, once my dummy load arrives I will be able to test for that. I should be able to use my Variac and set my PS to +/-75V and give you some idea of what it will do.
On another note. I have been reading about the Studio reference amp That Apex designed. It is supposed to work on +/-75V. You may want to look at that.
Blessings, Terry
Hi Guys
Considering that all of the Dx Blame variations are compromises of the original Blameless circuit, I would not bother with them. Certainly the fact there are board images available is enticing, but... I have not looked carefully at those to see what compromises have been made of the layout.
Again, I would strongly encourage you to look at the Leach Low TIM design and just expand it to the power you want. Cascoding is not needed to get more power TODAY. Just increase the supply values to +/-80V to +/-90V and use four pairs of output devices. Make sure all the active devices can tolerate the higher supply voltages.
Also note that just because your woofer is rated for 250W does not mean that it will tolerate that much power in the cabinet you are using. Speaker electrical ratings are for the voice coil only, and reflect use in a tight box with a lot of air resistance to cone movement. In a vented design, openback, or simply a larger sealed enclosure where the cone can move more freely, the amount of power required to reach full excursion is much reduced. This means the driver is effectively derated power-wise - maybe 100W pushes it to the excursion limit?
You say you've built some 50W or so amps. Do you have clip lights on these? Have you tried them with your intended cabinet load for the new high-power amp?
I think if you made some measurements to see how much power you use on average, the zeal for a lot of power might recede. On the other hand, it might show you that if you listen to dynamic music, that most of the time power is within that 50W window and just the peaks are being clipped. In that case, you should seriously consider making a class-G output stage.
In class-G, there are two supply levels, usually with the higher being twice the lower. The output stage looks superficially like a cascode but operates quite differently and does not have the plague of oscillations. This kind of amp is essentially identical to the Blameless up to the output stage. The same number of output BJTs is used as for the expanded Low TIM circuit, so eight devices per channel. You could build the Low TIM Leach, expanded to high voltage but with the class-G output.
The class-G output allows a reduction of the heat sink size and reduces the average waste heat from the amp. Idle conditions for the output stage are based on the lower supply rail value. The front-end and drivers must operate from the high rails for full output to be achieved.
Have fun
Kevin O'Connor
Considering that all of the Dx Blame variations are compromises of the original Blameless circuit, I would not bother with them. Certainly the fact there are board images available is enticing, but... I have not looked carefully at those to see what compromises have been made of the layout.
Again, I would strongly encourage you to look at the Leach Low TIM design and just expand it to the power you want. Cascoding is not needed to get more power TODAY. Just increase the supply values to +/-80V to +/-90V and use four pairs of output devices. Make sure all the active devices can tolerate the higher supply voltages.
Also note that just because your woofer is rated for 250W does not mean that it will tolerate that much power in the cabinet you are using. Speaker electrical ratings are for the voice coil only, and reflect use in a tight box with a lot of air resistance to cone movement. In a vented design, openback, or simply a larger sealed enclosure where the cone can move more freely, the amount of power required to reach full excursion is much reduced. This means the driver is effectively derated power-wise - maybe 100W pushes it to the excursion limit?
You say you've built some 50W or so amps. Do you have clip lights on these? Have you tried them with your intended cabinet load for the new high-power amp?
I think if you made some measurements to see how much power you use on average, the zeal for a lot of power might recede. On the other hand, it might show you that if you listen to dynamic music, that most of the time power is within that 50W window and just the peaks are being clipped. In that case, you should seriously consider making a class-G output stage.
In class-G, there are two supply levels, usually with the higher being twice the lower. The output stage looks superficially like a cascode but operates quite differently and does not have the plague of oscillations. This kind of amp is essentially identical to the Blameless up to the output stage. The same number of output BJTs is used as for the expanded Low TIM circuit, so eight devices per channel. You could build the Low TIM Leach, expanded to high voltage but with the class-G output.
The class-G output allows a reduction of the heat sink size and reduces the average waste heat from the amp. Idle conditions for the output stage are based on the lower supply rail value. The front-end and drivers must operate from the high rails for full output to be achieved.
Have fun
Kevin O'Connor
Last edited:
Struth, yes, i actually added clip lights to my 50w amp. They were indeed clipping only some peaks, but i have tested a 150w amp on the speakers. After almost frying eggs on the 150w amplifier radiator i realized how powerhungry my speakers are. Yes, the cabinets are vented but tuned and built for 300w woofers. Also, after pushing 150w through them (measured) they just crave for more. I am now driving them with a 40w Technics amp as i sold the others...
I don't want to go with a class g amp mainly because the secondary winding which i don't have.
I don't want to go with a class g amp mainly because the secondary winding which i don't have.
Hi Guys
The fact your 150W amp's heat sink got hot while driving the speaker is no reflection as to how power hungry the speakers are. That is waste heat. Every power amp produces waste heat.
The amount of waste heat varies with the signal level. The output device and speaker form a voltage divider where the division varies constantly. The current through the speaker produces sound while the same current through the output device makes heat. Ignoring idle heat and other issues and considering a pure sine wave, the worst-case heat output is when the voltage across the speaker equals that across the device, producing one-quarter of full output.
Odds are that the amp was running at around its 1/4 to 1/2 output level if it got very hot. The amp actually runs cooler above and below that level. This suggests that the power sent to the speakers in that "hot amp" condition was between about 38W and 75W.
Overall, your own experience might be telling you that you need more than 50W but actually much less than 150W - probably 100W will do.
Have fun
Kevin O'Connor
The fact your 150W amp's heat sink got hot while driving the speaker is no reflection as to how power hungry the speakers are. That is waste heat. Every power amp produces waste heat.
The amount of waste heat varies with the signal level. The output device and speaker form a voltage divider where the division varies constantly. The current through the speaker produces sound while the same current through the output device makes heat. Ignoring idle heat and other issues and considering a pure sine wave, the worst-case heat output is when the voltage across the speaker equals that across the device, producing one-quarter of full output.
Odds are that the amp was running at around its 1/4 to 1/2 output level if it got very hot. The amp actually runs cooler above and below that level. This suggests that the power sent to the speakers in that "hot amp" condition was between about 38W and 75W.
Overall, your own experience might be telling you that you need more than 50W but actually much less than 150W - probably 100W will do.
Have fun
Kevin O'Connor
Struth, i intended that to be a metaphor. I didn't use it as a heat quantifier, but more like a quantification of "i listened to it at max power" on not.
Another reason for building a amp this big is not only because i crave for max power. I can run it on the current amp for that matter. The idea is that these aren't the only speakers the amp is going to drive and also, home isn't the only place where it's gonna drive them. I used a 1X100w amp before and when i was at a party or such thing i was pushing it more and more till the driver popped. The protections came in and saved my speakers. Surprisingly, it wasn't distorting that bad (maaaax 1%). Luckily i had the Technics amp with me too just sitting there. I popped it in fast and got the music going in a matter of minutes.
I looked at the low tim amp but i think there's too much to modify to it.(more output devices have to be added; also, the small transistors need to be changed i think). I haven't done such thing never in my life. The only transistor work i did was a fan controller and some kind of motor power controller. Oh, and i also used some mosfets as switches, but nothing more.
I also searched the power trans.(MJE15031 and it's pair) from the superamp and saw they have only 50w of dissipation. Isn't that too low? I mean , others, like MJL4281 have much higher power dissipation of around 230w. Why is it like that?
I just realized those were the smaller trans.
Also, do i have to change the power trans. with modern ones? How do i do this?
Another reason for building a amp this big is not only because i crave for max power. I can run it on the current amp for that matter. The idea is that these aren't the only speakers the amp is going to drive and also, home isn't the only place where it's gonna drive them. I used a 1X100w amp before and when i was at a party or such thing i was pushing it more and more till the driver popped. The protections came in and saved my speakers. Surprisingly, it wasn't distorting that bad (maaaax 1%). Luckily i had the Technics amp with me too just sitting there. I popped it in fast and got the music going in a matter of minutes.
I looked at the low tim amp but i think there's too much to modify to it.(more output devices have to be added; also, the small transistors need to be changed i think). I haven't done such thing never in my life. The only transistor work i did was a fan controller and some kind of motor power controller. Oh, and i also used some mosfets as switches, but nothing more.
I also searched the power trans.(MJE15031 and it's pair) from the superamp and saw they have only 50w of dissipation. Isn't that too low? I mean , others, like MJL4281 have much higher power dissipation of around 230w. Why is it like that?
I just realized those were the smaller trans.
Also, do i have to change the power trans. with modern ones? How do i do this?
Last edited:
Hi Guys
I thought you said you had built a 50W amp and some other smaller ones? Maybe that was someone else in another thread?
So, you are really looking for a circuit and board that are ready to go right out of the box, so to speak.
The MJE15031/30 and MJE15034/35 are the driver transistors. These are a TO-220 case part that requires a small heat sink. The outputs are the MJL1302A/3281A, which bolt to the main heat sink and pass speaker current.
In general, changing old part numbers to new types is not problematic. Just assure that the new device has sufficient voltage, current and power rating for the position in the circuit. Newer devices often have higher gain and greater bandwidth, which are the attributes that may require tweaking of compensation or addition of extra compensation.
If you are envisioning running multiple speakers at the same time, so 4R loads or even lower, then you have to design the output stage for that condition. The 100W-8R amp if provided with enough current capability in the PSU and with sufficient parallel output transistors, could do 200W-4R, 400W-2R.. Each increase of output power causes more waste heat in the load. This requires bigger heat sinks and/or a provision for a thermally activated fan.
Even a crude voltage measurement while playing music as loud as you typically do will give some indication of the power used. It could save you a lot of time and money.
Sure, solid-state power is cheap these days, but in an absolute sense there are economies of construction, of time, and of operation that should be considered. For example, if you build the Super Leach with its 200W output or so, but you normally listen at SPLs no louder than it takes 1W to produce, then the waste power in the amp is far in excess of the signal requirement all the time and the operating cost is higher than it needs to be. The initial cost to build the 200W amp is also significantly higher than were you to build just a 5W amp, were the requirements assessed first.
With respect to waste heat mentioned above, I qualified the statement. To further qualify it: pure class-B bias, ie 180-degrees per circuit half, solid supply rails - no sag, pure resistive load.
In real life, power supplies sag, devices are biased not at zero current, maximum output power has to be a bit higher than rated to assure that all samples produce the rated output, the extra bit of supply headroom caused by accommodation of sag and of sample variance means waste heat is much higher than the ideal quoted. The one-third power preconditioning for testing assures everything in the chassis is at a worst-case temperature, so represents a worst-case performance condition. Marginally designed amps cannot withstand the preconditioning, and were that at 1/4 power instead, it would be no less punishing.
Have fun
Kevin O'Connor
I thought you said you had built a 50W amp and some other smaller ones? Maybe that was someone else in another thread?
So, you are really looking for a circuit and board that are ready to go right out of the box, so to speak.
The MJE15031/30 and MJE15034/35 are the driver transistors. These are a TO-220 case part that requires a small heat sink. The outputs are the MJL1302A/3281A, which bolt to the main heat sink and pass speaker current.
In general, changing old part numbers to new types is not problematic. Just assure that the new device has sufficient voltage, current and power rating for the position in the circuit. Newer devices often have higher gain and greater bandwidth, which are the attributes that may require tweaking of compensation or addition of extra compensation.
If you are envisioning running multiple speakers at the same time, so 4R loads or even lower, then you have to design the output stage for that condition. The 100W-8R amp if provided with enough current capability in the PSU and with sufficient parallel output transistors, could do 200W-4R, 400W-2R.. Each increase of output power causes more waste heat in the load. This requires bigger heat sinks and/or a provision for a thermally activated fan.
Even a crude voltage measurement while playing music as loud as you typically do will give some indication of the power used. It could save you a lot of time and money.
Sure, solid-state power is cheap these days, but in an absolute sense there are economies of construction, of time, and of operation that should be considered. For example, if you build the Super Leach with its 200W output or so, but you normally listen at SPLs no louder than it takes 1W to produce, then the waste power in the amp is far in excess of the signal requirement all the time and the operating cost is higher than it needs to be. The initial cost to build the 200W amp is also significantly higher than were you to build just a 5W amp, were the requirements assessed first.
With respect to waste heat mentioned above, I qualified the statement. To further qualify it: pure class-B bias, ie 180-degrees per circuit half, solid supply rails - no sag, pure resistive load.
In real life, power supplies sag, devices are biased not at zero current, maximum output power has to be a bit higher than rated to assure that all samples produce the rated output, the extra bit of supply headroom caused by accommodation of sag and of sample variance means waste heat is much higher than the ideal quoted. The one-third power preconditioning for testing assures everything in the chassis is at a worst-case temperature, so represents a worst-case performance condition. Marginally designed amps cannot withstand the preconditioning, and were that at 1/4 power instead, it would be no less punishing.
Have fun
Kevin O'Connor
Last edited:
Yes, i said i built a 50w amp and some smaller ones. The idea is that i had the schematic, which i assume it worked well and simply traced a pcb and corroded it. Installed the components there it goes.
About the smaller trans., the drivers, i edited my post like a minute after i submitted it saying i just realized these were the smaller trans. So yeah, the whole MJE15031 being the power trans. was actually a typo.
I am not envisioning running multiple speakers at the same time (and on the same channel). What i said is that this set of speakers i have now isn't the only pair of speakers i am going to use the amp with. They will be, most of the time, the ones used, but i do use others as well, depending on where i go and what equipment already exists there.
I also know that by paralleling the output trans. (or if they are in a larger number), the current capability increases and so the amp is going to be able to deliver much higher power (ideally twice the power for half the impedance) into lower impedances. I know i won't be getting exactly twice the power because the source, aka the PT, has some resistance to it and so works like one. As the current increases, the internal voltage drop becomes bigger and bigger, eventually lowering the rail voltage significantly and so reducing max output power at lower impedance. I know that there are also loses in circuitry that prevent the user from reaching that ideal 2X nominal power at 1/2 X nominal impedance.
Of course, as the current gets higher, the power that gets dissipated is higher, so yeah, i know i need more cooling.
And for god's sake, i get the fact that i waste power, time, money, resources by building a big amp if i listen to 1w worth of sound. BUT, for effin' god's sake, i do not do that. That's why i want to build a bigger amp.
Also, i know about the operating classes and so. Pure class B have great crossover distortion because of... well.. the crossover!
I hope i didn't bore you too much.
Back to amp schematics an typologies.
I don't think i'll manage to add more output trans. without first burning something and so... I can't really afford losing money now as i am in a vicious circle. I have a low budget to build an amp. I have a low budget because i couldn't go to play music at parties, so i didn't get paid. But also i can't do that because i don't have an am powerfull enough.
All in all, i'd rather make a freaking stereo 250w in 8 ohms amp using everything else i have lying around such as a 2X 55vac 1kw transformer, buffers, preamps and so on than building a 100w amp so i can make money so i can later upgrade it to a 250w amp because i know it's not going to happen. I know that after i build an amp i won't build another one only if the current one breaks. I am willing to pay for the best quality/performance/overall price ratio amp now so i can enjoy it for more time.
This is why i started this thread. I asked for your amiability and friendliness. I asked a question that many have asked and a few have answered. Which is the best not extremely complicated amp that can deliver 250w in 8 ohms worth of crystal clear sound? I have also been searching myself, BUT i do not have your vast experience and knowledge for which you have my entire respect.
Thank you for reading this much!
About the smaller trans., the drivers, i edited my post like a minute after i submitted it saying i just realized these were the smaller trans. So yeah, the whole MJE15031 being the power trans. was actually a typo.
I am not envisioning running multiple speakers at the same time (and on the same channel). What i said is that this set of speakers i have now isn't the only pair of speakers i am going to use the amp with. They will be, most of the time, the ones used, but i do use others as well, depending on where i go and what equipment already exists there.
I also know that by paralleling the output trans. (or if they are in a larger number), the current capability increases and so the amp is going to be able to deliver much higher power (ideally twice the power for half the impedance) into lower impedances. I know i won't be getting exactly twice the power because the source, aka the PT, has some resistance to it and so works like one. As the current increases, the internal voltage drop becomes bigger and bigger, eventually lowering the rail voltage significantly and so reducing max output power at lower impedance. I know that there are also loses in circuitry that prevent the user from reaching that ideal 2X nominal power at 1/2 X nominal impedance.
Of course, as the current gets higher, the power that gets dissipated is higher, so yeah, i know i need more cooling.
And for god's sake, i get the fact that i waste power, time, money, resources by building a big amp if i listen to 1w worth of sound. BUT, for effin' god's sake, i do not do that. That's why i want to build a bigger amp.
Also, i know about the operating classes and so. Pure class B have great crossover distortion because of... well.. the crossover!
I hope i didn't bore you too much.
Back to amp schematics an typologies.
I don't think i'll manage to add more output trans. without first burning something and so... I can't really afford losing money now as i am in a vicious circle. I have a low budget to build an amp. I have a low budget because i couldn't go to play music at parties, so i didn't get paid. But also i can't do that because i don't have an am powerfull enough.
All in all, i'd rather make a freaking stereo 250w in 8 ohms amp using everything else i have lying around such as a 2X 55vac 1kw transformer, buffers, preamps and so on than building a 100w amp so i can make money so i can later upgrade it to a 250w amp because i know it's not going to happen. I know that after i build an amp i won't build another one only if the current one breaks. I am willing to pay for the best quality/performance/overall price ratio amp now so i can enjoy it for more time.
This is why i started this thread. I asked for your amiability and friendliness. I asked a question that many have asked and a few have answered. Which is the best not extremely complicated amp that can deliver 250w in 8 ohms worth of crystal clear sound? I have also been searching myself, BUT i do not have your vast experience and knowledge for which you have my entire respect.
Thank you for reading this much!
Hi Guys
Your bolded question has been answered.
It is good that you know of all the details cited in your post.
The one detail I would like to emphasise again is simply that the schematic of a good 100W amp is the same as that of a good 200W amp. The only real difference is the supply voltages.
A PCB made for, say, a 100W Blameless amp will be the same PCB for your 250W amp. The only complication you may face is adding two more pairs of output transistors. However, this is just a matter of wiring those extra devices either hardwired or on "out-rigger" boards - etched or just perf doesn't matter.
You want simple, reliable and performance - the safest path is the Blameless amp. It requires no matched parts and has little if any DC offset. I believe in a previous post I listed what devices to use in each circuit portion (might have been a different thread?). The Leach amp variations do require some matching, so set those aside for this project. For that matter, most designs featuring complimentary symmetry from input to output require matching.
It is not rocket science despite some of the wild circuits people bandy about.
If you look at all the high-power amps presented on this forum and elsewhere, and all the commercially available amps for hifi from the past, 90% are of the type I described. The other 10% are far more complex than most people want to build and do not always provide any improvement in performance.
As a note about the Blameless amps: check out Doug Self's web site. He sells PCBs for the Blameless via his company, Signal Transfer Company.
Have fun
Kevin O'Connor
Your bolded question has been answered.
It is good that you know of all the details cited in your post.
The one detail I would like to emphasise again is simply that the schematic of a good 100W amp is the same as that of a good 200W amp. The only real difference is the supply voltages.
A PCB made for, say, a 100W Blameless amp will be the same PCB for your 250W amp. The only complication you may face is adding two more pairs of output transistors. However, this is just a matter of wiring those extra devices either hardwired or on "out-rigger" boards - etched or just perf doesn't matter.
You want simple, reliable and performance - the safest path is the Blameless amp. It requires no matched parts and has little if any DC offset. I believe in a previous post I listed what devices to use in each circuit portion (might have been a different thread?). The Leach amp variations do require some matching, so set those aside for this project. For that matter, most designs featuring complimentary symmetry from input to output require matching.
It is not rocket science despite some of the wild circuits people bandy about.
If you look at all the high-power amps presented on this forum and elsewhere, and all the commercially available amps for hifi from the past, 90% are of the type I described. The other 10% are far more complex than most people want to build and do not always provide any improvement in performance.
As a note about the Blameless amps: check out Doug Self's web site. He sells PCBs for the Blameless via his company, Signal Transfer Company.
Have fun
Kevin O'Connor
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.