Good morning, Shawn! 🙂
Congratulations!
A very nice test setup, I'm glad you've been successfull!
But with the quiescent current non-setup you've really been lucky. Before applying higher power into your dummy power resistors or a speaker, you should definitely adjust it accordingly in order to prevent thermal "runaway" at higher heat sink temperatures. In "short-term" test this could probably never ever happen, but under power and over hours or days it could just destroy most of your success in a matter of minutes.
I've read you have an oscilloscope? Adjustment to an optimum quiescent current yelds lowest crossover distortion (A/B switching) at the output; can easily be testet with a sinusoidal input and the scope at the output. Fine adjustment after rough ballpark setting with recommended current values, so to say...
Oh, and only nuts to bolt the transistors onto the heatsink, not a single ... 😀 Interresting 😀😀😀
Cheers,
Sebastian.
Congratulations!
A very nice test setup, I'm glad you've been successfull!
But with the quiescent current non-setup you've really been lucky. Before applying higher power into your dummy power resistors or a speaker, you should definitely adjust it accordingly in order to prevent thermal "runaway" at higher heat sink temperatures. In "short-term" test this could probably never ever happen, but under power and over hours or days it could just destroy most of your success in a matter of minutes.
I've read you have an oscilloscope? Adjustment to an optimum quiescent current yelds lowest crossover distortion (A/B switching) at the output; can easily be testet with a sinusoidal input and the scope at the output. Fine adjustment after rough ballpark setting with recommended current values, so to say...
Oh, and only nuts to bolt the transistors onto the heatsink, not a single ... 😀 Interresting 😀😀😀
Cheers,
Sebastian.
I just shaved and now I'm looking at...
I have 1/2 amp fuses on the PCB. Is that a get out of jail card? I would never risk the entire gig baby!
Have you been up to what I was last night cause that comes close but is so far away. Detail it for me man! This DIY and the kids are all reading what we say and do? I don't get a word of what you are saying, though I guess, I get the drift.
Shawn.
sek said:
I have 1/2 amp fuses on the PCB. Is that a get out of jail card? I would never risk the entire gig baby!
Have you been up to what I was last night cause that comes close but is so far away. Detail it for me man! This DIY and the kids are all reading what we say and do? I don't get a word of what you are saying, though I guess, I get the drift.
Dude, I have no idea what you're are saying but the vibes sound very good 🙂
Shawn.
Whoa....
I leave the thread for what like 12 hours and .....
Congratulations Shawn on a smoke-less activation. It's good to hear that your happy with the thing so far. Thanks also to other members for your assistance.
Shawn; I have emailed a file that contains the latest schematics and layouts for the amp and other stuff plus a brief guide on the setup procedure. This is copied below for your (and others) convenience.
Cheers
Quasi
________________________________
Setup Guide
The setup is done without a load connected to the power amp module. First check your work and make sure output devices are insulated from heatsink.
1. Remove fuses and replace with 100 ohm 5 watt resistors.
2. Connect amp to power supply and turn on. Make sure power supply polarity is correct.
3. Check output offset voltage and adjust VR1 to achieve an offset of less than 10 mV.
4. All being well a place a voltmeter across one of the 100 ohm resistors.
5. Adjust VR2 to set the output stage bias current, by measuring the voltage across one of the 100 ohm resistors. For the 6 FET board set for a voltage of 9 volts. This equates to a bias current of 30mA per FET pair or 90 mA total. For the 10 FET board set for a voltage of 15 volts.
6. All being well replace the 100 ohm resistors with 10 ohms 1 watt resistors and re-adjust VR2 to get 0.9 volts for the 6 FET board or 1.5 volts for the 10 FET board.
7. Once done, remove the resistors and put the fuses in. Re-check the offset voltage and adjust with VR1 if necessary. The amp is ready.
I leave the thread for what like 12 hours and .....
Congratulations Shawn on a smoke-less activation. It's good to hear that your happy with the thing so far. Thanks also to other members for your assistance.
Shawn; I have emailed a file that contains the latest schematics and layouts for the amp and other stuff plus a brief guide on the setup procedure. This is copied below for your (and others) convenience.
Cheers
Quasi
________________________________
Setup Guide
The setup is done without a load connected to the power amp module. First check your work and make sure output devices are insulated from heatsink.
1. Remove fuses and replace with 100 ohm 5 watt resistors.
2. Connect amp to power supply and turn on. Make sure power supply polarity is correct.
3. Check output offset voltage and adjust VR1 to achieve an offset of less than 10 mV.
4. All being well a place a voltmeter across one of the 100 ohm resistors.
5. Adjust VR2 to set the output stage bias current, by measuring the voltage across one of the 100 ohm resistors. For the 6 FET board set for a voltage of 9 volts. This equates to a bias current of 30mA per FET pair or 90 mA total. For the 10 FET board set for a voltage of 15 volts.
6. All being well replace the 100 ohm resistors with 10 ohms 1 watt resistors and re-adjust VR2 to get 0.9 volts for the 6 FET board or 1.5 volts for the 10 FET board.
7. Once done, remove the resistors and put the fuses in. Re-check the offset voltage and adjust with VR1 if necessary. The amp is ready.
Yeah Shawn,
just stick to what Quasi said, that's a way more comprehensive and understandable explanation than I could give. 😉
Unfortunately, transistors blow/melt much quicker than fuses. Thermal "runaway" describes the case where one of the output devices heats up much more than the others. In that case, the hot one draws much more current than the others and melts. This does not neccessarily mean that the total current draw increases, it just shares unevenly. That's why the rail fuse still feels safe while some FETs could already be burning. Only after the meltdown does the fuse recognize the problem. 🙁
On the other hand, 1/2 amp fuses limit your total power draw to about 30 Watts. So, as I posted earlier, you won't recognize problems due to bad biasing during your short-term low power tests...
Don't worry, just adjust everything like Quasi said. 🙂
I thouht I could prevent to write ... "not a single ... screwhead" 😀😀😀
Well, it's easy: if you have an oscilloscope and a signal generator, just connect them to the amplifier module, apply a sinusoidal input, watch for distortion (spike) free output and enjoy.
If you don't have an oscilloscope and a signal generator, then don't . 😉
You could then (as an option) do that later using your computer, as there is good signal generator and oscilloscope reading software available for free.
Sebastian.
just stick to what Quasi said, that's a way more comprehensive and understandable explanation than I could give. 😉
Unfortunately, transistors blow/melt much quicker than fuses. Thermal "runaway" describes the case where one of the output devices heats up much more than the others. In that case, the hot one draws much more current than the others and melts. This does not neccessarily mean that the total current draw increases, it just shares unevenly. That's why the rail fuse still feels safe while some FETs could already be burning. Only after the meltdown does the fuse recognize the problem. 🙁
On the other hand, 1/2 amp fuses limit your total power draw to about 30 Watts. So, as I posted earlier, you won't recognize problems due to bad biasing during your short-term low power tests...
Don't worry, just adjust everything like Quasi said. 🙂
I thouht I could prevent to write ... "not a single ... screwhead" 😀😀😀
Well, it's easy: if you have an oscilloscope and a signal generator, just connect them to the amplifier module, apply a sinusoidal input, watch for distortion (spike) free output and enjoy.
If you don't have an oscilloscope and a signal generator, then don't . 😉
You could then (as an option) do that later using your computer, as there is good signal generator and oscilloscope reading software available for free.
Sebastian.
The drivers heatsink;
The drivers heatsink will run very warm (hot) and considering the dissipation of the 4 devices this is normal. But don't worry the transistors here are pretty tough. Anyone building high power versions of this amp should employ a case fan to replace the internal air, this will help reduce the temperature of the drivers (and other stuff).
With respect to whether 30mA bias per FET is enough, I personally think it is and anything more is really a case of quickly diminishing returns. But there is no reason why the bias can't be turned up higher, and as Jacco has pointed out can be overbiased to destruction.
But anything more than 75mA - 100mA per FET pair is just a waste with this design and then maybe you should consider class A. Be aware though that this level of biasing will heat things up a bit.
Finally, if anyone would like the file I've sent to Shawn, send an email.
Cheers
The drivers heatsink will run very warm (hot) and considering the dissipation of the 4 devices this is normal. But don't worry the transistors here are pretty tough. Anyone building high power versions of this amp should employ a case fan to replace the internal air, this will help reduce the temperature of the drivers (and other stuff).
With respect to whether 30mA bias per FET is enough, I personally think it is and anything more is really a case of quickly diminishing returns. But there is no reason why the bias can't be turned up higher, and as Jacco has pointed out can be overbiased to destruction.
But anything more than 75mA - 100mA per FET pair is just a waste with this design and then maybe you should consider class A. Be aware though that this level of biasing will heat things up a bit.
Finally, if anyone would like the file I've sent to Shawn, send an email.
Cheers
Now that you're asking, I would be glad to receive it.
(Please use the mail function below or just the address that you sent the schematic and layout file to.)
Thanks,
Sebastian.
(Please use the mail function below or just the address that you sent the schematic and layout file to.)
Thanks,
Sebastian.
Hi Sebastian,
It seems you already have it, I have sent you the latest version just in case.
Cheers
It seems you already have it, I have sent you the latest version just in case.
Cheers
Hi Guys,
Can someone please comment this clipping indicator curcuit for my quasi amp?It's on the professor Leach website and it's used for his leach amp clipping indicator.
http://users.ece.gatech.edu/~mleach/lowtim/graphics/clip.pdf
Thanks in advance...
Miodrag Pejin
P.S. I'm really enojying the amp!!!
Can someone please comment this clipping indicator curcuit for my quasi amp?It's on the professor Leach website and it's used for his leach amp clipping indicator.
http://users.ece.gatech.edu/~mleach/lowtim/graphics/clip.pdf
Thanks in advance...
Miodrag Pejin
P.S. I'm really enojying the amp!!!
Re: The drivers heatsink;
Click Here for a copy
I'm surprised more of these amps have not been built. Folks, it’s a winner. I'll be back in a few days with some case action. Thanks everybody for the guidance.
Shawn.
Click Here for a copy
I'm surprised more of these amps have not been built. Folks, it’s a winner. I'll be back in a few days with some case action. Thanks everybody for the guidance.
Shawn.
Re: Re: The drivers heatsink;
Shawn,
Although I have not built quasi's amp yet, nor have I built an amp for some years, I would 120% agree with you that quasi's amp is excellent. My view is based on the process quasi took to end up with a solid design, the uncanny way quasi chose parts for the quasi store that has shown how flexible the amp is to alternate parts, the comments on the design, the many positive comments of those that have built the amp and their ears happiness .... (raching out the collection tray).
I think many are into the "fashionable" class A amps, ignoring arguments to other amp classes aside, seems to be one of the many unfair reasons there has not been as much interest in quasi's wonderful amp. As we all know fashionable does not mean better.
I am curious at some point what the differences might be to quasi's n-channle MOSFET amp and the BiPolar varient might be? The other fashionable sides, tubes of course aside, are the BiPolar fans and the CHIP AMP fans. The latter is popular for many due to the simple and low part count appeal to many first time or impatient builders needing instant gratification!
quasi's design has a low part count, little in way of capacitors in the signal path, excellent thinking to the PCB design and a friendly and helpful kind gent of quasi when people run into problems. Remember this is a hobby for quasi as well and as quasi so well indicated the quasi formula of time for his day a number of posts back, it is likley just as well this is not a intense or many first time builders attracted to quasi's amp. quasi only has so much time.
I looked at the Holton n-challen amp, which many were happy with in sound, it was just very fussy about parts or setting it up. Lars Zeta was interesting, but in my opinion had, in my humble mind, throry of concern. There were a couple of smaller n-channel MSOFET amps, but just not clean enough for my needs. There were some complementary designs well regarded, like the Holton AV400, but for me personally the P-Channel devices were rather limiting in power handling. BiPolars were of some interest to me, but availability just to build a couple amps too limiting or a sea of fake devices to swim through while MOSFETs were easy to find and flexible to subsitute for one's power, audio, price or availability.
So yes, you Shawn like the other's that have completed this amp before you know the gem of a amp quasi has brought to us. I know I am really looking forward to when I have all the parts and heatsinks to build my first couple modules to give a listen as well. Then I can work on case and secondary support elements and circuits I wish to incorporate into my amps around the amp module. My goal is build these amps with the expectation I will not likely need to build any other amp design again, except try out a tube design and one Class A for tweeters for curiousity. quasi's amp will be my base amps for the years ahead is my expectation. This is why I am taking the time to research and select parts that will last and do best job for price possible.
I for one would like to give quasi a long round of applause and in lieu of quasi's time maybe the quasi store can be stocked up with more parts 😉
Regards,
John L. Males
Willowdale, Ontario
Canada
21 June 2006 19:19
TomWaits said:
Shawn,
Although I have not built quasi's amp yet, nor have I built an amp for some years, I would 120% agree with you that quasi's amp is excellent. My view is based on the process quasi took to end up with a solid design, the uncanny way quasi chose parts for the quasi store that has shown how flexible the amp is to alternate parts, the comments on the design, the many positive comments of those that have built the amp and their ears happiness .... (raching out the collection tray).
I think many are into the "fashionable" class A amps, ignoring arguments to other amp classes aside, seems to be one of the many unfair reasons there has not been as much interest in quasi's wonderful amp. As we all know fashionable does not mean better.
I am curious at some point what the differences might be to quasi's n-channle MOSFET amp and the BiPolar varient might be? The other fashionable sides, tubes of course aside, are the BiPolar fans and the CHIP AMP fans. The latter is popular for many due to the simple and low part count appeal to many first time or impatient builders needing instant gratification!
quasi's design has a low part count, little in way of capacitors in the signal path, excellent thinking to the PCB design and a friendly and helpful kind gent of quasi when people run into problems. Remember this is a hobby for quasi as well and as quasi so well indicated the quasi formula of time for his day a number of posts back, it is likley just as well this is not a intense or many first time builders attracted to quasi's amp. quasi only has so much time.
I looked at the Holton n-challen amp, which many were happy with in sound, it was just very fussy about parts or setting it up. Lars Zeta was interesting, but in my opinion had, in my humble mind, throry of concern. There were a couple of smaller n-channel MSOFET amps, but just not clean enough for my needs. There were some complementary designs well regarded, like the Holton AV400, but for me personally the P-Channel devices were rather limiting in power handling. BiPolars were of some interest to me, but availability just to build a couple amps too limiting or a sea of fake devices to swim through while MOSFETs were easy to find and flexible to subsitute for one's power, audio, price or availability.
So yes, you Shawn like the other's that have completed this amp before you know the gem of a amp quasi has brought to us. I know I am really looking forward to when I have all the parts and heatsinks to build my first couple modules to give a listen as well. Then I can work on case and secondary support elements and circuits I wish to incorporate into my amps around the amp module. My goal is build these amps with the expectation I will not likely need to build any other amp design again, except try out a tube design and one Class A for tweeters for curiousity. quasi's amp will be my base amps for the years ahead is my expectation. This is why I am taking the time to research and select parts that will last and do best job for price possible.
I for one would like to give quasi a long round of applause and in lieu of quasi's time maybe the quasi store can be stocked up with more parts 😉
Regards,
John L. Males
Willowdale, Ontario
Canada
21 June 2006 19:19
Re: Re: Re: Re: The drivers heatsink;
"Cheques" as in plural??? Yippie!!!!!
Quasi quick question re C7 feedback capacitor:
I understood from my prior question you indicated any voltage is likely ok for this cap, even as low as 16V. I seem to recall your schematic had listed the voltage as 50V. The question is what is the likely typical voltage the capacitor may see and what in theory is the maximum voltage the capacitor will see (VRail, 2XVrail?)? My prior understanding was in all likelyhood this capacitor will blow so the voltage rating was not that critical.
Regards,
John L. Males
Willowdale, Ontario
Canada
22 June 2006 07:34
quasi said:
"Cheques" as in plural??? Yippie!!!!!
Quasi quick question re C7 feedback capacitor:
I understood from my prior question you indicated any voltage is likely ok for this cap, even as low as 16V. I seem to recall your schematic had listed the voltage as 50V. The question is what is the likely typical voltage the capacitor may see and what in theory is the maximum voltage the capacitor will see (VRail, 2XVrail?)? My prior understanding was in all likelyhood this capacitor will blow so the voltage rating was not that critical.
Regards,
John L. Males
Willowdale, Ontario
Canada
22 June 2006 07:34
Hi John,
The DC voltage at the junction of R17 & R18 is 0v plus any output offset error (< 20mV). This therefore should be the same for C7.
If the amp goes faulty then anything could happen, including C7 trying to test the physical boundaries of whatever enclosure you have your modules in.
You could use a capacitor rated at least the same as your supply rails to survive fault conditions but what would be the point? The amp would still be broken.
I have a 25v capacitor in my amp if that gives you comfort.
Cheers
The DC voltage at the junction of R17 & R18 is 0v plus any output offset error (< 20mV). This therefore should be the same for C7.
If the amp goes faulty then anything could happen, including C7 trying to test the physical boundaries of whatever enclosure you have your modules in.
You could use a capacitor rated at least the same as your supply rails to survive fault conditions but what would be the point? The amp would still be broken.
I have a 25v capacitor in my amp if that gives you comfort.
Cheers
Update on my own amp;
The amp I built ( http://www.diyaudio.com/forums/showthread.php?postid=586001#post586001 ) has been running now for about 15 months without any signs of trouble and has proven to be a very stable design. As Shawn suggests it delivers effortless power and control over the audio spectrum. But what has pleased me the most is how quiet it is without signal.
So with careful layout and tidy wiring, constructors should end up with as good or even better result.
Cheers
The amp I built ( http://www.diyaudio.com/forums/showthread.php?postid=586001#post586001 ) has been running now for about 15 months without any signs of trouble and has proven to be a very stable design. As Shawn suggests it delivers effortless power and control over the audio spectrum. But what has pleased me the most is how quiet it is without signal.
So with careful layout and tidy wiring, constructors should end up with as good or even better result.
Cheers
Hi Quasi,
Thanks for answer. I asked just in case it seemed practical to choose a "safe" value. Seems for all intents not practical. I will go with the best cap I can find to use there as was suggested I find best electrolytic can.
It was explained I need to keep the Time Constant of C7/R18 to 100ms. That was pointed out as I tried to find alternate uF values I could actually purchase. The challenge was being able to fit a C7 onto the PCB location without me doing more GIMP work to modify the board with the voltages I thought or were availabel for me to buy.
Your answer will enable me to have greater choice to choose from to find a C7. I know I am not likely to hear a difference using a Nippon, Elna, Sprague, Richey, Rubycon, Panasonic, or etc, but one place I found sells based on value/voltage rating and does not care about make. Given I can go for the make that in theory would be best choice for the function with no difference in cost I will do so. I seem to recall Sprague's have a low ESR, as do the Rubycons. I am still not sure of the temperment of the Rubycon, so I will likely go for a Sprague.
BTW, I found a Linux program that could almost allow me to input your amp schematic with a GUI. The intent was to do simulations to find the voltages and currents with my lack of understanding some of the special math outside Ohm's Law I could not figure out. It was suppose to use a SPICE based back end to do the simulation. Darn it all, it would not allow me to flip the transistors like T4 to face the base in, so I am stalled again as I try to find a program that will allow me to to input the schematic and then do a simulation. I only want to check some voltages for a few parts and thought be great to have the values as well in case I run into odd problems to enable me to at least find the points of difference. If I understood the few non Ohm's Law exceptions I would be just fine to create a spreadsheet to provide me the values. I will keep trying to find a program. It will not stop me from building the amp once I have all the parts I need. I just thought be helpful in some parts selection, especially if I cannot find enough MJE340/MJE350 that I need (very tough device that has very few, if any equals in ruggedness), to enable me to try out a few alternates to T8 and implement the R6 value change for the supply rail voltage I would be using, agmonst tuning some support circuits like VU meter, clipping, et al.
Take care and thanks again,
John L. Males
Willowdale, Ontario
Canada
22 June 2006 10:08
Thanks for answer. I asked just in case it seemed practical to choose a "safe" value. Seems for all intents not practical. I will go with the best cap I can find to use there as was suggested I find best electrolytic can.
It was explained I need to keep the Time Constant of C7/R18 to 100ms. That was pointed out as I tried to find alternate uF values I could actually purchase. The challenge was being able to fit a C7 onto the PCB location without me doing more GIMP work to modify the board with the voltages I thought or were availabel for me to buy.
Your answer will enable me to have greater choice to choose from to find a C7. I know I am not likely to hear a difference using a Nippon, Elna, Sprague, Richey, Rubycon, Panasonic, or etc, but one place I found sells based on value/voltage rating and does not care about make. Given I can go for the make that in theory would be best choice for the function with no difference in cost I will do so. I seem to recall Sprague's have a low ESR, as do the Rubycons. I am still not sure of the temperment of the Rubycon, so I will likely go for a Sprague.
BTW, I found a Linux program that could almost allow me to input your amp schematic with a GUI. The intent was to do simulations to find the voltages and currents with my lack of understanding some of the special math outside Ohm's Law I could not figure out. It was suppose to use a SPICE based back end to do the simulation. Darn it all, it would not allow me to flip the transistors like T4 to face the base in, so I am stalled again as I try to find a program that will allow me to to input the schematic and then do a simulation. I only want to check some voltages for a few parts and thought be great to have the values as well in case I run into odd problems to enable me to at least find the points of difference. If I understood the few non Ohm's Law exceptions I would be just fine to create a spreadsheet to provide me the values. I will keep trying to find a program. It will not stop me from building the amp once I have all the parts I need. I just thought be helpful in some parts selection, especially if I cannot find enough MJE340/MJE350 that I need (very tough device that has very few, if any equals in ruggedness), to enable me to try out a few alternates to T8 and implement the R6 value change for the supply rail voltage I would be using, agmonst tuning some support circuits like VU meter, clipping, et al.
Take care and thanks again,
John L. Males
Willowdale, Ontario
Canada
22 June 2006 10:08
Hi Key.
that NFB cap can be low voltage when the amp is working correctly.
You measure the Vac across the cap when it is working hard at low frequency.
If a significant voltage develops across it then it has too high an ESR at that frequency. This is quite a rigorous test, as suggested by D.Self and others. The diode across the cap is to prevent a highish reverse voltage developing across the electrolytic if very low frequency signals drive the output in that direction.
The one area that the electrolytic will struggle is excessive DC on the output. Here the resistor ratio does not divide due to the open circuit set by the NFB cap and it feels the full brunt of either the +ve or -ve offset. Bang! but again inverse parallel diodes avoid the blow up. Just check the dissipation of the feedback resistor (small is beautiful in this case) it will fuse. Whereas normally you require little or no temperature cycling in this resistor.
that NFB cap can be low voltage when the amp is working correctly.
You measure the Vac across the cap when it is working hard at low frequency.
If a significant voltage develops across it then it has too high an ESR at that frequency. This is quite a rigorous test, as suggested by D.Self and others. The diode across the cap is to prevent a highish reverse voltage developing across the electrolytic if very low frequency signals drive the output in that direction.
The one area that the electrolytic will struggle is excessive DC on the output. Here the resistor ratio does not divide due to the open circuit set by the NFB cap and it feels the full brunt of either the +ve or -ve offset. Bang! but again inverse parallel diodes avoid the blow up. Just check the dissipation of the feedback resistor (small is beautiful in this case) it will fuse. Whereas normally you require little or no temperature cycling in this resistor.
Andrew,
Thanks for the info regarding the NFB cap/resistor. This should mean when I follow quasi's initial amp set up for dc offset nulling and bias setup via VR1 and VR2 I will have low risk to blowing C7 as part of nulling the DC offset as low as possible.
Thanks for the insight how to get a feel if the ESR is too high based on frequency.
As FYI I had followed up with some research regarding the Time Constant of C7/R18 you gave me guidance on. I did two things. I tried C7 values of 47uF, 68uF and 120uF to find the matching R18 to keep the Time Constant to 100ms as you suggested using a web based calculator for the purpose. These were all at the 63.2/37.8 percentages. I then tried for 90% charge time (I am not sure if this implied 10% discharge) for all those C7 values as well as for 100uF to see what the resistor values would be. I know many amps use a 100uF/1K combination so I have decided to not tinker with these values now that I have now found a source for some 100uF caps that will fit the space on the PCB. I do not know enough to understand the effect of the different combinations beyond the percentage, effect of the different values or percentages to "customize", say for a tweeter, midrange or (sub)woofer amp should it be useful.
Regards,
John L. Males
Willowdale, Ontario
Canada
22 June 2006 11:45
Thanks for the info regarding the NFB cap/resistor. This should mean when I follow quasi's initial amp set up for dc offset nulling and bias setup via VR1 and VR2 I will have low risk to blowing C7 as part of nulling the DC offset as low as possible.
Thanks for the insight how to get a feel if the ESR is too high based on frequency.
As FYI I had followed up with some research regarding the Time Constant of C7/R18 you gave me guidance on. I did two things. I tried C7 values of 47uF, 68uF and 120uF to find the matching R18 to keep the Time Constant to 100ms as you suggested using a web based calculator for the purpose. These were all at the 63.2/37.8 percentages. I then tried for 90% charge time (I am not sure if this implied 10% discharge) for all those C7 values as well as for 100uF to see what the resistor values would be. I know many amps use a 100uF/1K combination so I have decided to not tinker with these values now that I have now found a source for some 100uF caps that will fit the space on the PCB. I do not know enough to understand the effect of the different combinations beyond the percentage, effect of the different values or percentages to "customize", say for a tweeter, midrange or (sub)woofer amp should it be useful.
Regards,
John L. Males
Willowdale, Ontario
Canada
22 June 2006 11:45
Hi Quasi, and everyone following this thread,
Lots of you have been building these amps from parts scavenged / recycled from old equipment. A few of you have said the MJE340/MJE350 transistors are hard to find.
As luck would have it I have a LARGE stock of MJE340 / MJE350 equivalent transistors. These are made by Fairchild, and are KSE340 / KSE350. As far as I know they are identical to the motorola devices.
If anyone is interested, I'm happy to send you a small quantity of these for free in the interests of DIY goodwill!
If there is a lot of interest, maybe one person from each country could handle distribution to their fellows, I really don't want to send out a hundred envelopes..
Email me offline, lets keep this thread on track.
thanks,
Len.
Lots of you have been building these amps from parts scavenged / recycled from old equipment. A few of you have said the MJE340/MJE350 transistors are hard to find.
As luck would have it I have a LARGE stock of MJE340 / MJE350 equivalent transistors. These are made by Fairchild, and are KSE340 / KSE350. As far as I know they are identical to the motorola devices.
If anyone is interested, I'm happy to send you a small quantity of these for free in the interests of DIY goodwill!
If there is a lot of interest, maybe one person from each country could handle distribution to their fellows, I really don't want to send out a hundred envelopes..
Email me offline, lets keep this thread on track.
thanks,
Len.