Aleph formulas

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
SY said:
Loop gain is quite a standard term. I first ran across it studying Crowhurst maybe 30 years ago. AFAIK, he wasn't a god, just a fine writer and engineer. And all too mortal.


It wouldn't hurt for everyone to should run a google search before posting here. I remember you pointing that out to me once...(though I had already done it)
Save time and embarassment since so many people are out to get you. ;)
 
Seems to me that "loop gain" is a pretty generic and vague term. It gives no indication of the scope of the loop or the condition within which the loop is being operated. As far as protecting newbies, it is best to employ more descriptive or specific terms to avoid people interpreting them based strictly on their own experiences, which can lead to miscommunication and spawn further confusion. Not everyone is well versed in the widespread literature of electronics nor possesses an extensive library on the subject for reference.
 
Reading this thread gives me many kind of feelings. First I'm getting more and more excited+honoured as Mr.Pass answers more questions that I asked. I agree that maybe he have the everybody's respect more than I imagine.
Just reading the page3 of this thread, smiling because Mr.Pass post a reply, and laughing when reading the post of GRollins. It is true that only GRollins answers all my questions systematicly (I get the right answers). But I feel sorry, because he wrote that someone(s) else is only post to attack his answers, not answering the original question. Even words from the dictionary becoming the target, looking at the following thread.
Anyway, GRollins, I'm sure behind all that there is a Friendship spirit behind all the critics. Maybe it is not realized that all those who always attacks your answers, realizing it or not, have high appreciation for your knowledge. Critics written is like a chance to find a little gap to say "I'm as good as you are". That means you are good, right?
 
Thanks for the new (to me) term, Jocko.

Grey,
I didn't know that the gain would also increase with the current increase (or at least I didn't think about it). Interesting.
Still, it seems that you're trying to get around the original point of what I was trying to say. You don't need more input voltage from the preamp. As Fred said, the feedback resistors largely determine the closed-loop gain of the amp. Also, what would be the point of decreasing the resistors if you aren't going to rebias the front end to correctly bias the output stage?

Even though Fred and Jocko tend to post with a little bit of a hostile attitude, from what I see the criticism is usually deserved. The points they make are based on fact. If you talk out your @ss, they are going to call you on it. It actually helps differentiate those who know from those who like to post.
 
Grataku,

Still standing up for Grey? He went on the attack and as usual he made some mistakes so I brought attention to the fact.

Like noboby special said.

If you talk out your @ss, they are going to call you on it.

He set himself as Guru and I have yet to see him retract any of his incorrect statements.

The mention of both jackals and Mt. Olympus might cause a bit of confusion. Mt. Olympus is associated with Greek mythology. Zeus was the boss for the Greek gods. Anubis, a god of the dead and depicted as a man with the head of a jackal like animal was a god in Egyptian mythology. Of course Hermanubis, was a god who combined Anubis and Greek god Hermes and was popular during the Roman Empire.

Jocko and Fred bristle at the comparisons to gods and are quite happy with their status as mere mortals.

Regards,
Jam
 
One of the side effects of having an 8 month old child is that I have even less time to spend online than I used to. Those who wish to take my non-presence online as indicating anything ought to try getting as much done in a day as I do before attempting to make an issue of it.
Jocko,
I happily concede the existence of the term "loop gain," though I'm not sure how much utility it has in day-to-day useage if authors such as H&H don't find it of sufficient interest to put it in the index. (Read what I said carefully--I said it wasn't in the index. Reading skills are on the decline here in the US, but that's another topic for another day.) It may be in the text for all I know, but if out of three or four books that I looked through yesterday, not one had it listed, then it's not going to supplant open loop, closed loop, and NFB in common usage, even if it's technically correct.
Search the web? Sometimes I do. Sometimes I don't. In this case it might have provided evidence of existence of the term (I haven't tried the search yet), but I maintain that if the mainstream texts don't care to list it in the index, that's good enough for me.
I regard TI application notes and such as being much more authoritative than the run of the mill web nonsense, but confound their hides, they rarely, if ever, have an index. Yes, more recent versions of Acrobat have a search feature, but that still begs the question of common usage. At any rate, Jocko, I have that particular TI paper around here somewhere, and I'll try to find time to look at your reference in the next day or two.
In any event, one point to Jocko for existence of the term loop gain.
Jam,
Your rather specious posts make it difficult to reply, except in kind. I'll do what I can, however.
--I didn't 'quote' H&H. I simply noted that the term "loop gain" didn't appear in the index.
--For what it's worth, the title is Gray's Anatomy, not Grey. Of the two spellings, Gray is by far the more common. Makes it difficult to get my name spelled correctly. (Thanks for getting my name right. Long story as to how I ended up with an unusual name. Too long to go into here.) Needless to say, I get a lot of Greg, Gary, etc. Oh, well. At least no one confuses me with other authors, so I guess it works to my benefit in at least that one way.
--Operating environment for active devices. Hmmm. This could get interesting. Are you saying that unbypassed resistance under the cathode/Source/emittor doesn't provide local degenerative feedback? Used to be so for tubes. The last I knew, it was so for solid state as well.
--'He set himself as Guru...' Curious, that one. Not once have I ever claimed guru status for myself. Nor fame, nor EE degree, etc. (Double major in geology/psychology, though I took a number of electronics courses in high school and college. Not enough to qualify as expert. Never claimed such. I was actually more interested in physics at the time; the electronics was just for fun.) If you can find me a post where I said such a thing (other than in jest), I'd be much obliged. The closest thing I can think of would be cases such as Fred (he was Harry Haller then) claiming that the Aleph-X would oscillate based on simulations and his EE experience and whatever else. The problem with that was that I had a functioning circuit at my elbow. In my book, reality beats theory/simulations every day of the week. (If you feel that Fred's word should win the day in spite of the fact that there are who knows how many functioning Aleph-X amps out there, then I'm not sure what to say.) Ditto for whether the current draw in an Aleph-X summed to DC on each rail and so forth. He was wrong, time and again. Not a convincing argument for seeking an EE degree, in my view. Incidentally, you speak of me not retracting incorrect statements, yet I haven't seen you say as much to Fred. Hyprocrisy, perhaps?
All this side-taking reminds me of nothing so much as the whole Democrat/Republican thing in US politics. Clinton spent private time with Monica then tried to weasel out it and the Republicans went on the attack; Democrats on the defensive. No middle ground. Now Bush has himself a war and the Democrats are on the attack and the Republicans are on the defensive. No middle ground. Personally, I find the whole thing repugnant. If all that energy could be tapped, we'd get something done. Instead, there's a lot of churning going on. Same here. I post. Fred doesn't like it, even though he ends up saying pretty much the same thing I said, though in different words. Maybe he doesn't like my phrasing. If you look at our posts, you'll find that I don't follow Fred around, posting after him. Quite the reverse, in fact. Pity he can't just use all that energy answering questions. Suppose he'd simply answered the posts earlier in this thread. I wouldn't have had to say anything at all. Cool idea, huh?
Any chance of it happening?
Don't bet on it.

Grey

P.S.: I'm sure there are skeighty-eight other points I haven't addressed, but I've already spent much more time online than I should have and I need to go check on Olivia. Anything else will have to wait.
Oh, wait, one thing occurs to me. Someone said something above about the front end in an Aleph and don't you pretty much have to increase the current when you reduce the load resistance. Taken as such, yes. However, as I recall (might be wrong, and the post was far enough upstream that it's not available on my screen at the moment) Lumanauw originally postulated changing the load resistor. Only. The idea of increasing the front end bias to compensate came later (Me? Lumanauw? Someone else? I don't remember). On the one hand, you can look at this as a pragmatic thing. If you want the amp to continue functioning after changing the load resistor, you're going to need to do something. The easiest thing to do is increase bias current in the front end. On the other hand, as originally stated, changing the load resistor in isolation would have certain effects, among them lowered gain, etc. In proper experimental fashion, you change only one thing at a time, then note the results. In an Aleph, you have to change at least two things, which tends to muddy the waters. Was the change due to increased current or lowered resistance? Or both? A philosophical point, perhaps, but it might be one that would cause some confusion.
 
Jam,
I am not sticking up for Grey or any others and I am not trying to stick it to you, Fred or Jocko. I am trying to be friend with everybody although there is probably no way to succed in this very confrontational climate.

Yes, Grey gives a "hint" of wanting to portray himself as a "guru", his posts maybe long, and may contain a certain amount of minor BS but don't they all? The question is how many cartoons and and snide remarks do we need to see per every Grey's posts?
 
Grey,

Maybe it is a time for a truce among all involved..........

I would hope that if any of us make an error we will own up to it. As for me not taking Fred to task..........please read some of my earlier posts.

I even built his current source (bad word) and it worked, a lot better than some of the fancier ones that were posted. I am more interested in that correct information gets out and that newbies understand the danger of electricity. This is ment to be a fun hobby.

Grataku,

If you have read some some of my earlier posts I have always poked (prodded) fun at Jocko, Fred and even Steve Eddy. And as for you..............:D :D :D :D ...............no one is safe!

Regards,
Jam
 
Grey:

There is no need to concede anything. This is not a contest. However, many of my texts do list "loop gain" in the index.

From my standpoint, you seem to try to make things more simple than they really are, and in the process, this leads to confusion.

Use of misleading terminology is a chronic problem on this forum. I judge that by the amount of e-mails I receive asking for clarification of things that seem to baffle people on most any given day.

I felt it would have been easier in this case to head off a series of incoming questions by clearing things up beforehand.

I can not comment about the circuit in question, as the last time that I looked at one of Nelson's designs, it had "Forte" on the front panel.

"Hey, Jocko.....what do you think of this new amp Nelson designed? Aren't these the same output transistors that you are using? Does **** ******* know about these yet? Are they any good?" End of story, and poking my nose into Nelson's work.

Jocko
 
diyAudio Retiree
Joined 2002
Convincing argument for seeking an EE degree?

From lumanauw: http://www.diyaudio.com/forums/showthread.php?postid=247696#post247696



"One more question. In second stage, there are various placement of 1nF stabilizer capacitor. In aleph design it places in lower G-D of mosfet,but in volksamp, it is on the top (on the constant current G-S ofmosfet). Which is the right one, Mr Pass?"

The compensation capacitor on the top circuit is for the stability of the dynamic current source. On most of the Alephs it is either 0.047uF from the collector to emitter of the bipolar NPN in that part of the circuit, or is 1000nF from the collector to base of the NPN. The compensation capacitor for the voltage gain around the amp is from the output to the junction of the 392 load resistor from the gate resistors for the output transistors and varies in value for the different Aleph models. The Alephs with 100K ohm feedback resistors also include compensation with 10pF across this resistor. I believe Grey used the same scheme as the A30 and A60 only
the 10pF but with no compensation on
the NPNs in the dynamic current sources. I will of course leave it to him
to explain his design choices, which I
am not aware of the reasons for.

http://www.diyaudio.com/forums/showthread.php?postid=249304#post249304


Addressing the explanation for the source resistors I offer another view point as I can't quite follow Mr. Rollins but he might be saying some similar but in terms that just don't quite click for me (and perhaps a few others) when reading it.


http://www.diyaudio.com/forums/showthread.php?postid=249255#post249255

" One, it provides a current sense reading for the current-limiting circuit (the lower NPN and the voltage divider at its base). Two, it gives the output MOSFET an operating environment more similar to the upper MOSFET (the Aleph current source). By making the environment more similar, the two MOSFETs behave more similarly, reducing distortion by some small fraction. (For that matter, there's some local feedback involved, which will also reduce distortion.) And most importantly, three, it tends to damp out differences between devices operating in parallel."

The current limiting could be addressed in other ways as well. For example clamping the gate to source voltage for the mosfets with a zener (since the gate to source voltage determines the mosfet current) or sensing the current trough the sense resistors or a drain resistor on the same mosfet, which would not influence the transconductance like a source degeneration resistor would. Designing the current limit circuit after including the source resistor for another reason seems a more likely explanation since it is a good place to sense the current through the output stage. I will not claim to know the reason for doing the protection circuit like this as Mr. Pass is obviously much more familiar with the design goals than anyone else will ever be.

To claim to understand the design choices and seemingly the order in which they were made for this part of the circuit is much further out on a limb than some of us would like to sit.

"Making the environment more similar" for the devices is such a broad term that it seems to me that it could mean about anything. I would point out that one device is used as a transconductance stage (voltage to current) in the open loop gain of the amp and the other as a dynamic current source, sensing the output current in the amp and including a BJT transistor in a local feedback loop. The AC voltage across one mosfet is increasing while decreasing across the other. The AC current is different through one mosfet than the other for this topology. I just can't understand what "making the environment more similar" means and underscores the point Jocko makes about using engineering terms. They don't have to be complicated and starting with terms like voltage, current, power, and temperature would have given us someplace to start, in trying to understand this.

"It tends to damp out differences between devices operating in parallel" again gives me little clue as the meaning of this. Damp and damping as engineering terms get me little closer to understand the point. This just too general to me to dispute or confirm due to the seeming lack of real substance in the statement.

I would like to offer another reason that seems more likely for the inclusion of source resistors. As the current increases the voltage across the source resistor increases. For a given input voltage (from the gate to the end off the source resistor not attached to the mosfet) the gate to source voltage has to decrease as the voltage across the source resistor increases. This tends to give a stable DC current trough the mosfet, aiding in DC offset stability and equal power sharing for devices in parallel. I hope this is straight forward and I believe it didn't require the use of any engineering terms that were at all obscure.

As for other allegations and slurs, you will have to reference the actual post for me to respond. I would like to note that the tolerance for abuse in name calling is at an all time low and my post
http://www.diyaudio.com/forums/showthread.php?postid=248520#post248520

that seemed to prompt your emotional responces contained neither, but was based on responses that were though out and not meant to inflame any one. To disagree with someone statements is not the same as an attack on there character. Please keep that in remind when making responses to differing viewpoints than yours. You are bringing more disrespect on yourself than anyone else could, even if they were making the effort to do that and nothing else.

Fred
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.