Some of the Randy Slone designs such as the one sometimes called the "11-4" are virtually Self designs. In fact it's a sort of generic design that a lot of people have made variations on. I personally built a couple of "11-4's" and think they are quite good.
Some here have made the basic "blamless" but I don't know if anyone has actually built the "load invariant" or "class G'. You can get PCBs for these but they are a bit pricey. However, considering the complexity it may be warranted.
Self like CFB output sections which although they can have considerable benefits also can be troublesome to get right (i.e., non oscillating). The physical PCB layout can contribute to stability or lack thereof and even Self admits the problems with parasitics are not well understood. The above mentioned 11-4 required me to re-do the PCB layout 3 or 4 times before I got it behaving well. I have a suspicion in this regard that older, slower (low fT) output devices such as MJ15003 may be less troublesome than newer higher speed (higer fT) devices but I'm not willing to make a broad definitive claim to this effect.
Some here have made the basic "blamless" but I don't know if anyone has actually built the "load invariant" or "class G'. You can get PCBs for these but they are a bit pricey. However, considering the complexity it may be warranted.
Self like CFB output sections which although they can have considerable benefits also can be troublesome to get right (i.e., non oscillating). The physical PCB layout can contribute to stability or lack thereof and even Self admits the problems with parasitics are not well understood. The above mentioned 11-4 required me to re-do the PCB layout 3 or 4 times before I got it behaving well. I have a suspicion in this regard that older, slower (low fT) output devices such as MJ15003 may be less troublesome than newer higher speed (higer fT) devices but I'm not willing to make a broad definitive claim to this effect.
Hi Sam.
" I have a suspicion in this regard that older, slower (low fT) output devices such as MJ15003 may be less troublesome than newer higher speed (higer fT) devices "
Yes it's always good to have a dominant pole in any loop be it global or local. I sold thousands of amplifiers in the '70's using MJ15003/4 CFP and CFT's without problems once loop gain limiting resistors were inserted usually in the driver emitter leg.
Cheerrs,
greg
" I have a suspicion in this regard that older, slower (low fT) output devices such as MJ15003 may be less troublesome than newer higher speed (higer fT) devices "
Yes it's always good to have a dominant pole in any loop be it global or local. I sold thousands of amplifiers in the '70's using MJ15003/4 CFP and CFT's without problems once loop gain limiting resistors were inserted usually in the driver emitter leg.
Cheerrs,
greg
Yes, there are comments about Doctor Self amplifiers.
You can find them here:
http://www.diyaudio.com/forums/showthread.php?postid=450093#post450093
And also here, as the thread was splitted:
http://www.diyaudio.com/forums/showthread.php?postid=451093#post451093
regards,
Carlos
You can find them here:
http://www.diyaudio.com/forums/showthread.php?postid=450093#post450093
And also here, as the thread was splitted:
http://www.diyaudio.com/forums/showthread.php?postid=451093#post451093
regards,
Carlos
Regarding the "Blameless" design: you will find a certain amount of criticsm of this design (although those who have built them are often happy with them). I think this misses the point. It seems obvious to me that there was no intent to equate the term "blamless" with "optimal" or "perfect" etc. The intent was to design a fairly basic, generic design with no glaring errors or falts in order to use it as a baseline for for analysing the effect of various distortion reducing techniques.
For example you don't *HAVE* to have a bypass cap on the input sections CCS but most people think it is a good idea. Self used his blameless design to investigate the assumed benefits of this cap with different CCS schemes and to determine what effectiveness of different cap values are. Using the blamless design let him isolate just those effects.
For example you don't *HAVE* to have a bypass cap on the input sections CCS but most people think it is a good idea. Self used his blameless design to investigate the assumed benefits of this cap with different CCS schemes and to determine what effectiveness of different cap values are. Using the blamless design let him isolate just those effects.
Hi,
Well, very nice answers, but none deep technical, i built the Class A, that is a little different then the class AB blameless, and simulated it as others did, is fine, the THD over 1KHz is one of
the lowest can be found the bass is strong and a lot of highs and don't have to 'cook it' in order to settle with the temp. is stable in the firsts seconds.
What I think, despite all the critics that is a little not clear in a very special range of mids compared to a custom tubes pwr amp S.E. made by a friend.
In any case I see that no one built this one, the class A , I raised up the voltage rallies to about +/-28 Vdc and it drives more power, works good with my ESL 63.
Probably I have to compare it to one more serious.
Regards Williams
Well, very nice answers, but none deep technical, i built the Class A, that is a little different then the class AB blameless, and simulated it as others did, is fine, the THD over 1KHz is one of
the lowest can be found the bass is strong and a lot of highs and don't have to 'cook it' in order to settle with the temp. is stable in the firsts seconds.
What I think, despite all the critics that is a little not clear in a very special range of mids compared to a custom tubes pwr amp S.E. made by a friend.
In any case I see that no one built this one, the class A , I raised up the voltage rallies to about +/-28 Vdc and it drives more power, works good with my ESL 63.
Probably I have to compare it to one more serious.
Regards Williams
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.