mwh-eng said:
the "pop-lization" of spice is that even a person without much knowledge can quickly confirm complicated circuits works. In the good old days, you simply have to give into "industry veterans" because there are potentially countless factors at work in a circuit. and you would have no way of knowing if someone is BSing you.
With spice, you can confirm that quickly.
That's not to say that spice is the do-all and be-all. But it does make it a lot easier to figure out who is a BS artist.
"pop-lization" of spice
"But it does make it a lot easier to figure out who is a BS artist."
I am not having ANY problem figuring this out at all, only counting how many of them have been involved in this!
I am having a hard time trying not to wet my pants laughing at this point. This is put on right? How can anyone trust someone who can't read a data sheet to interpret Spice modeling results........
"But it does make it a lot easier to figure out who is a BS artist."
I am not having ANY problem figuring this out at all, only counting how many of them have been involved in this!
I am having a hard time trying not to wet my pants laughing at this point. This is put on right? How can anyone trust someone who can't read a data sheet to interpret Spice modeling results........
janneman said:
is it possible, Jan, that the regulator doesn't work the way you think it does?
janneman said:
I don't know, Jan. I thought a regulator is there to kee ot the ripples. can you find a regulator that leaves ripples in place?
janneman said:
that might be a nice business: to make regulators that have lot of ripples,
janneman said:
well, if a voltage follower (or whatever) gets the job done, why should we care about the name?
Re: "pop-lization" of spice
I don't know. My trouble is even one level lower: how can anyone trust someone who cannot read a datasheet, period?
and how can anyone trust someone who tries to get his way out of a mistake by claiming that the device designers implemented it wrong? or by confusing everyone and their mothers hoping to divert the discussion.
The soft start circuit, afterall, is right there for us to see, right? Maybe some of us need eye-glasses,, before typing fingers.
Fred Dieckmann said:[BHow can anyone trust someone who can't read a data sheet to interpret Spice modeling results........
I don't know. My trouble is even one level lower: how can anyone trust someone who cannot read a datasheet, period?
and how can anyone trust someone who tries to get his way out of a mistake by claiming that the device designers implemented it wrong? or by confusing everyone and their mothers hoping to divert the discussion.
The soft start circuit, afterall, is right there for us to see, right? Maybe some of us need eye-glasses,
, before typing fingers.Re: Back to the origional question.........
My reasons for starting this thread here were the following, in decreasing order of importance:
And just to set the record straight, I was not looking for a slow start or slow turn-on circuit, though I now realise that it might be a nifty fringe benefit if I can get it. Nothing in Randy's book, or in my email exchanges with him, ever mentioned the term "slow start" or anything like it.
I doubt today that this thread's discussion particularly cares what my original priorities were, but I just thought I'd put things down in writing, since Fred asked. Have fun, though it'd be nice if I got some usable circuits/circuit ideas out of it too.
Millwood, it seems you're the only person who's actually built this circuit on veroboard. Does it work for my requirements? Is there any fundamental flaw?
Well, since you ask me, I'll tell you what was on my mind. I saw the circuit in Randy Slone's book, and thought it would be useful as a general-purpose preamp PSU. I had some private correspondence with Randy, and he confirmed that this is a high performance circuit and works well. We didn't get into the details of what "high performance" meant.Fred Dieckmann said:Jan......... the only mysterious reading I did was reading tcpip asking if this circuit could improve the regulator performance. I don't think he was looking for a slow start circuit but was looking for a circuit to improve the regulation. I will admit most of the post here are off to a VERY SLOW START to providing him any help..... I think it is fine that people can find a slow start circuit in the data sheet, but that is not the circuit function tcpip is looking for. Is it tcpip?
My reasons for starting this thread here were the following, in decreasing order of importance:
- To get a confirmation from you guys whether this circuit will work as shown to give me a usable PSU
- To get some inputs about how the performance of this circuit will be, i.e. can it be improved with small changes here and there?
- To get a deeper understanding of how it works, so that I may be able to suggest some improvements myself, and
- To get some inputs and pointers about better regulated PSU schematics. I've seen the Super Regulator schematic on Per-Ander's Website; I want something with about one-tenth that parts count, if possible.
- inexpensive (about USD 10-15 max for all components for a +/- PSU)
- each supply rail's components must fit into a 3.2"x4" PCB, if not both supply rails
- easily available parts where I stay. (I can relax this constraint for an exceptional circuit which meets all other requirements, and order from Digikey.)
- output voltage settable using presets in the range 5-25V
- output current upto 500mA per rail
- very low output impedance
- very high rejection of perturbations in input voltage
- very steady output voltage irrespective of perturbations in load impedance
- very low noise on the output rail
- low temperature- and time- drift of parameters
- infinite heat dissipation
And just to set the record straight, I was not looking for a slow start or slow turn-on circuit, though I now realise that it might be a nifty fringe benefit if I can get it. Nothing in Randy's book, or in my email exchanges with him, ever mentioned the term "slow start" or anything like it.
I doubt today that this thread's discussion particularly cares what my original priorities were, but I just thought I'd put things down in writing, since Fred asked.
Have fun, though it'd be nice if I got some usable circuits/circuit ideas out of it too. Millwood, it seems you're the only person who's actually built this circuit on veroboard. Does it work for my requirements? Is there any fundamental flaw?
Even though millwoodn't, I would........
"Well, since you ask me, I'll tell you what was on my mind. "
Thanks, I don't think it too much reading between the lines to see what you were after. The Zetex ZVP3306 (a ZVP 3310 will work as well) are available from Digikey. Feel free to Email me with any questions on the circuit. Maybe you can get some answers without having to sort through the hysterics here. I do think that some simple enhancements to improve the regulator are well worthwhile and am sorry this thread got off to such a Slow Start
Post Script for jocko
I wonder if that floor wax works on milledwood floors or if he has gotten his just desserts?
"Well, since you ask me, I'll tell you what was on my mind. "
Thanks, I don't think it too much reading between the lines to see what you were after. The Zetex ZVP3306 (a ZVP 3310 will work as well) are available from Digikey. Feel free to Email me with any questions on the circuit. Maybe you can get some answers without having to sort through the hysterics here. I do think that some simple enhancements to improve the regulator are well worthwhile and am sorry this thread got off to such a Slow Start
Post Script for jocko
I wonder if that floor wax works on milledwood floors or if he has gotten his just desserts?
Re: Re: Back to the origional question.........
I am not sure of your requirements as I don't have very sophisticated test and measurement equipment. All I can tell is that it does provide a soft start (not completely tho.). I fed it with an unregulated DC supply so there wasn't a whole lot of fluctuation to begin with. But the output was rock-solid, not even fluctuate by 1mv.
tcpip: one thing I forgot to mention is that the ripple rejection does improve with the "soft start" circuit. I have some simulated numbers from home and if you want I can post it later.
Yes, Fred, it is a soft start circuit,. No Fred, the guys at National did not implement it wrong. Yes, Fred, your "invention" is simply a carbon copy of their "incorrectly implemented" circuit.
tcpip said:
I am not sure of your requirements as I don't have very sophisticated test and measurement equipment. All I can tell is that it does provide a soft start (not completely tho.). I fed it with an unregulated DC supply so there wasn't a whole lot of fluctuation to begin with. But the output was rock-solid, not even fluctuate by 1mv.
tcpip said:
tcpip: one thing I forgot to mention is that the ripple rejection does improve with the "soft start" circuit. I have some simulated numbers from home and if you want I can post it later.
Yes, Fred, it is a soft start circuit,
. No Fred, the guys at National did not implement it wrong. Yes, Fred, your "invention" is simply a carbon copy of their "incorrectly implemented" circuit.
millwood said:
Could Millwood, not janneman, have been right all along?
Hi Millwood,
Well, I have to admit, after doing some more headscratching, that I probably wasn't right about that cap. Let me give you my reasoning, see what you make of it.
If the output let's say goes up, and the delta between output and adj is constant (1.25V I guess), and there is impedance in the adj leg, the adj goes up with the same level as the output. That doesn't bring the output back to what it is supposed to be. There's no regulation here. So far so good.
Now we put a cap on the adj, so the jump in output level is absorbed (shorted to gnd) so the adjust stays as it is, and the output tends to go down back to where it was.
Agreed?
Jan Didden
Since I was the one who brought that up, I found a very clear graph in the datasheets which tell me impulse load regulation improves with a cap. Like Millwood, I tend to think the datasheet authors know their stuff, so I guess that's how it works. It's visually confusing, since it really looks like a low pass filter in the feedback loop. Just goes to show you can't always trust instinct...
Rune
Rune
"tcpip: one thing I forgot to mention is that the ripple rejection does improve with the "soft start" circuit. I have some simulated numbers from home and if you want I can post it later."
It does?............... how does it do that, since the transistor turns off when the cap charges up to the voltage at the ADJ terminal and the transistor turns off. I guess you want to split hairs (and you do) the RC in parallel with the programing resistor to ground does provide some improvement but not much since the R in the RC delay is much larger than resistor from the ADJ terminal to ground.
Thanks for proving the common theory that Spice is of very limited use if you don't have a good grasp of how the circuit is supposed to work.
Don't stop by all means though, we are having too much fun! You might want to post the Spice model for the regulator since the typical models for things like op amps and regulators are pretty simplified.
Here are some Spice models but I have not played with it to see how accurate they are. It is a bit more detailed than a macromodel though and comparisons of model results with the data sheet graphs would be good a starting point.
*SRC=LM317MOT;LM317MOT;Voltage Reg.;Motorola;
*SYM=LM317MOT
.SUBCKT LM317MOT 1 2 3
*Connections Input Adj. Output
*LM317A voltage regulator - Motorola
J1 1 3 4 JN
Q2 5 5 6 QPL .1
Q3 5 8 9 QNL .2
Q4 8 5 7 QPL .1
Q5 81 8 3 QNL .2
Q6 3 81 10 QPL .2
Q7 12 81 13 QNL .2
Q8 10 5 11 QPL .2
Q9 14 12 10 QPL .2
Q10 16 5 17 QPL .2
Q11 16 14 15 QNL .2
Q12 3 20 16 QPL .2
Q13 1 19 20 QNL .2
Q14 19 5 18 QPL .2
Q15 3 21 19 QPL .2
Q16 21 22 16 QPL .2
Q17 21 3 24 QNL .2
Q18 22 22 16 QPL .2
Q19 22 3 241 QNL 2
Q20 3 25 16 QPL .2
Q21 25 26 3 QNL .2
Q22A 35 35 1 QPL 2
Q22B 16 35 1 QPL 2
Q23 35 16 30 QNL 2
Q24A 27 40 29 QNL .2
Q24B 27 40 28 QNL .2
Q25 1 31 41 QNL 5
Q26 1 41 32 QNL 50
D1 3 4 DZ
D2 33 1 DZ
D3 29 34 DZ
R1 1 6 310
R2 1 7 310
R3 1 11 230
R4 1 17 120
R5 1 18 5.6K
R6 4 8 125K
R7 8 81 135
R8 10 12 12.4K
R9 9 3 190
R10 13 3 3.6K
R11 14 3 5.8K
R12 15 3 110
R13 20 3 5.1K
R14 2 24 12.5K
R15 24 241 2.4K
R16 16 25 6.7K
R17 16 40 12K
R18 30 41 160
R19 16 31 170
R20 26 27 6.8K
R21 27 40 510
R22 3 41 200
R23 33 34 13K
R24 28 29 105
R25 28 32 4
R26 32 3 .1
C1 21 3 30PF
C2 21 2 30PF
C3 25 26 5PF
CBS1 5 3 2PF
CBS2 35 3 1PF
CBS3 22 3 1PF
.MODEL JN NJF(BETA=1E-4 VTO=-7)
.MODEL DZ D(BV=6.3)
.MODEL QNL NPN(EG=1.22 BF=80 RB=100 CCS=1.5PF TF=.3NS TR=6NS CJE=2PF
+ CJC=1PF VAF=100)
.MODEL QPL PNP(BF=40 RB=20 TF=.6NS TR=10NS CJE=1.5PF CJC=1PF VAF=50)
.ENDS
*LM317 TI voltage regulator - pin order: In, Adj, Out
*TI adjustable voltage regulator pkg:TO-3
.SUBCKT XLM317 1 2 3 **Changes my be required on this line**
J1 1 3 4 JN
Q2 5 5 6 QPL .1
Q3 5 8 9 QNL .2
Q4 8 5 7 QPL .1
Q5 81 8 3 QNL .2
Q6 3 81 10 QPL .2
Q7 12 81 13 QNL .2
Q8 10 5 11 QPL .2
Q9 14 12 10 QPL .2
Q10 16 5 17 QPL .2
Q11 16 14 15 QNL .2
Q12 3 20 16 QPL .2
Q13 1 19 20 QNL .2
Q14 19 5 18 QPL .2
Q15 3 21 19 QPL .2
Q16 21 22 16 QPL .2
Q17 21 3 24 QNL .2
Q18 22 22 16 QPL .2
Q19 22 3 241 QNL 2
Q20 3 25 16 QPL .2
Q21 25 26 3 QNL .2
Q22A 35 35 1 QPL 2
Q22B 16 35 1 QPL 2
Q23 35 16 30 QNL 2
Q24A 27 40 29 QNL .2
Q24B 27 40 28 QNL .2
Q25 1 31 41 QNL 5
Q26 1 41 32 QNL 50
D1 3 4 DZ
D2 33 1 DZ
D3 29 34 DZ
R1 1 6 310
R2 1 7 310
R3 1 11 190
R4 1 17 82
R5 1 18 5.6K
R6 4 8 100K
R7 8 81 130
R8 10 12 12.4K
R9 9 3 180
R10 13 3 4.1K
R11 14 3 5.8K
R12 15 3 72
R13 20 3 5.1K
R14 2 24 12K
R15 24 241 2.4K
R16 16 25 6.7K
R17 16 40 12K
R18 30 41 130
R19 16 31 370
R20 26 27 13K
R21 27 40 400
R22 3 41 160
R23 33 34 18K
R24 28 29 160
R25 28 32 3
R26 32 3 .1
C1 21 3 30PF
C2 21 2 30PF
C3 25 26 5PF
CBS1 5 3 2PF
CBS2 35 3 1PF
CBS3 22 3 1PF
.MODEL JN NJF(BETA=1E-4 VTO=-7)
.MODEL DZ D(BV=6.3)
.MODEL QNL NPN(EG=1.22 BF=80 RB=100 CCS=1.5PF TF=.3NS TR=6NS CJE=2PF
+ CJC=1PF VAF=100)
.MODEL QPL PNP(BF=40 RB=20 TF=.6NS TR=10NS CJE=1.5PF CJC=1PF VAF=50)
.ENDS XLM317 **changes may be required on this line**
*SRC=LM337K;LM337K;Voltage Reg.;;
*SYM=LM317TI
.SUBCKT LM337K 8 1 19
*Connections Input Adj. Output
*LM337 negative voltage regulator
.MODEL QN NPN (BF=50 TF=1N CJC=1P)
.MODEL QPOUT PNP (BF=50 TF=1N RE=.2 CJC=1P)
.MODEL QP PNP CJC=1P TF=2N
.MODEL DN D
.MODEL D2 D BV=12 IBV=100U
R10 25 6 1K
Q3 8 17 16 QPOUT
Q4 8 25 17 QP
R18 19 17 250
R19 19 16 .3
G1 8 6 1 18 .1
C7 6 2 .04U
R24 2 8 100
I_ADJ 0 1 65U
R26 8 25 200K
Q5 25 4 19 QP
R27 16 4 200
R28 7 4 7K
D1 8 7 D2
D2 8 6 DN
V1 18 19 1.25
.ENDS
Some 317data sheets are at
http://katalogi.iele.polsl.gliwice.pl/en/ enter LM317 is search part symbol field
And as a heads up:
From:
http://www.pcbcafe.com/BOOKS/SpiceHandBook/05_chapter04-02.php
"SPICE tip: Interestingly, there are three models of the LM117 in the Intusoft model library. One model gave the correct DC output voltage and the correct bode response. One model gave an incorrect DC output voltage but the correct bode response, and one model didn’t converge. Surprisingly, all three models are transistor level. This is just another example of the necessity of testing previously unused models against their datasheet performance in order to ensure model accuracy. Incidentally, the model used in these simulations is the LM317TI model."
Some more useful ap notes info:
http://www.national.com/pf/LM/LM317.html
Have I left anything out millwood?
It does?............... how does it do that, since the transistor turns off when the cap charges up to the voltage at the ADJ terminal and the transistor turns off. I guess you want to split hairs (and you do) the RC in parallel with the programing resistor to ground does provide some improvement but not much since the R in the RC delay is much larger than resistor from the ADJ terminal to ground.
Thanks for proving the common theory that Spice is of very limited use if you don't have a good grasp of how the circuit is supposed to work.
Don't stop by all means though, we are having too much fun! You might want to post the Spice model for the regulator since the typical models for things like op amps and regulators are pretty simplified.
Here are some Spice models but I have not played with it to see how accurate they are. It is a bit more detailed than a macromodel though and comparisons of model results with the data sheet graphs would be good a starting point.
*SRC=LM317MOT;LM317MOT;Voltage Reg.;Motorola;
*SYM=LM317MOT
.SUBCKT LM317MOT 1 2 3
*Connections Input Adj. Output
*LM317A voltage regulator - Motorola
J1 1 3 4 JN
Q2 5 5 6 QPL .1
Q3 5 8 9 QNL .2
Q4 8 5 7 QPL .1
Q5 81 8 3 QNL .2
Q6 3 81 10 QPL .2
Q7 12 81 13 QNL .2
Q8 10 5 11 QPL .2
Q9 14 12 10 QPL .2
Q10 16 5 17 QPL .2
Q11 16 14 15 QNL .2
Q12 3 20 16 QPL .2
Q13 1 19 20 QNL .2
Q14 19 5 18 QPL .2
Q15 3 21 19 QPL .2
Q16 21 22 16 QPL .2
Q17 21 3 24 QNL .2
Q18 22 22 16 QPL .2
Q19 22 3 241 QNL 2
Q20 3 25 16 QPL .2
Q21 25 26 3 QNL .2
Q22A 35 35 1 QPL 2
Q22B 16 35 1 QPL 2
Q23 35 16 30 QNL 2
Q24A 27 40 29 QNL .2
Q24B 27 40 28 QNL .2
Q25 1 31 41 QNL 5
Q26 1 41 32 QNL 50
D1 3 4 DZ
D2 33 1 DZ
D3 29 34 DZ
R1 1 6 310
R2 1 7 310
R3 1 11 230
R4 1 17 120
R5 1 18 5.6K
R6 4 8 125K
R7 8 81 135
R8 10 12 12.4K
R9 9 3 190
R10 13 3 3.6K
R11 14 3 5.8K
R12 15 3 110
R13 20 3 5.1K
R14 2 24 12.5K
R15 24 241 2.4K
R16 16 25 6.7K
R17 16 40 12K
R18 30 41 160
R19 16 31 170
R20 26 27 6.8K
R21 27 40 510
R22 3 41 200
R23 33 34 13K
R24 28 29 105
R25 28 32 4
R26 32 3 .1
C1 21 3 30PF
C2 21 2 30PF
C3 25 26 5PF
CBS1 5 3 2PF
CBS2 35 3 1PF
CBS3 22 3 1PF
.MODEL JN NJF(BETA=1E-4 VTO=-7)
.MODEL DZ D(BV=6.3)
.MODEL QNL NPN(EG=1.22 BF=80 RB=100 CCS=1.5PF TF=.3NS TR=6NS CJE=2PF
+ CJC=1PF VAF=100)
.MODEL QPL PNP(BF=40 RB=20 TF=.6NS TR=10NS CJE=1.5PF CJC=1PF VAF=50)
.ENDS
*LM317 TI voltage regulator - pin order: In, Adj, Out
*TI adjustable voltage regulator pkg:TO-3
.SUBCKT XLM317 1 2 3 **Changes my be required on this line**
J1 1 3 4 JN
Q2 5 5 6 QPL .1
Q3 5 8 9 QNL .2
Q4 8 5 7 QPL .1
Q5 81 8 3 QNL .2
Q6 3 81 10 QPL .2
Q7 12 81 13 QNL .2
Q8 10 5 11 QPL .2
Q9 14 12 10 QPL .2
Q10 16 5 17 QPL .2
Q11 16 14 15 QNL .2
Q12 3 20 16 QPL .2
Q13 1 19 20 QNL .2
Q14 19 5 18 QPL .2
Q15 3 21 19 QPL .2
Q16 21 22 16 QPL .2
Q17 21 3 24 QNL .2
Q18 22 22 16 QPL .2
Q19 22 3 241 QNL 2
Q20 3 25 16 QPL .2
Q21 25 26 3 QNL .2
Q22A 35 35 1 QPL 2
Q22B 16 35 1 QPL 2
Q23 35 16 30 QNL 2
Q24A 27 40 29 QNL .2
Q24B 27 40 28 QNL .2
Q25 1 31 41 QNL 5
Q26 1 41 32 QNL 50
D1 3 4 DZ
D2 33 1 DZ
D3 29 34 DZ
R1 1 6 310
R2 1 7 310
R3 1 11 190
R4 1 17 82
R5 1 18 5.6K
R6 4 8 100K
R7 8 81 130
R8 10 12 12.4K
R9 9 3 180
R10 13 3 4.1K
R11 14 3 5.8K
R12 15 3 72
R13 20 3 5.1K
R14 2 24 12K
R15 24 241 2.4K
R16 16 25 6.7K
R17 16 40 12K
R18 30 41 130
R19 16 31 370
R20 26 27 13K
R21 27 40 400
R22 3 41 160
R23 33 34 18K
R24 28 29 160
R25 28 32 3
R26 32 3 .1
C1 21 3 30PF
C2 21 2 30PF
C3 25 26 5PF
CBS1 5 3 2PF
CBS2 35 3 1PF
CBS3 22 3 1PF
.MODEL JN NJF(BETA=1E-4 VTO=-7)
.MODEL DZ D(BV=6.3)
.MODEL QNL NPN(EG=1.22 BF=80 RB=100 CCS=1.5PF TF=.3NS TR=6NS CJE=2PF
+ CJC=1PF VAF=100)
.MODEL QPL PNP(BF=40 RB=20 TF=.6NS TR=10NS CJE=1.5PF CJC=1PF VAF=50)
.ENDS XLM317 **changes may be required on this line**
*SRC=LM337K;LM337K;Voltage Reg.;;
*SYM=LM317TI
.SUBCKT LM337K 8 1 19
*Connections Input Adj. Output
*LM337 negative voltage regulator
.MODEL QN NPN (BF=50 TF=1N CJC=1P)
.MODEL QPOUT PNP (BF=50 TF=1N RE=.2 CJC=1P)
.MODEL QP PNP CJC=1P TF=2N
.MODEL DN D
.MODEL D2 D BV=12 IBV=100U
R10 25 6 1K
Q3 8 17 16 QPOUT
Q4 8 25 17 QP
R18 19 17 250
R19 19 16 .3
G1 8 6 1 18 .1
C7 6 2 .04U
R24 2 8 100
I_ADJ 0 1 65U
R26 8 25 200K
Q5 25 4 19 QP
R27 16 4 200
R28 7 4 7K
D1 8 7 D2
D2 8 6 DN
V1 18 19 1.25
.ENDS
Some 317data sheets are at
http://katalogi.iele.polsl.gliwice.pl/en/ enter LM317 is search part symbol field
And as a heads up:
From:
http://www.pcbcafe.com/BOOKS/SpiceHandBook/05_chapter04-02.php
"SPICE tip: Interestingly, there are three models of the LM117 in the Intusoft model library. One model gave the correct DC output voltage and the correct bode response. One model gave an incorrect DC output voltage but the correct bode response, and one model didn’t converge. Surprisingly, all three models are transistor level. This is just another example of the necessity of testing previously unused models against their datasheet performance in order to ensure model accuracy. Incidentally, the model used in these simulations is the LM317TI model."
Some more useful ap notes info:
http://www.national.com/pf/LM/LM317.html
Have I left anything out millwood?
regulation characteristic
I am still sticking to my simplified brain-model.
Please do not misunderstand me wrong, I love PSpice.
But normally I make up my own brain model first and then
check the opinion of Spice....
Happines comes when both match somehow AND
reality proves it in the end. (Sometimes reality does
not care about my thoughts... )
My simple model brain model of the L317 is the following:
-Floating regulator
- It regulates the output 1.2V above the adjustment pin,
no matter on which potential both are.
-Internal structure can be modeled by a floating
band gap reference (or any other precise reference) followed
by an power OP amp (unfortunately without much sinking capability) which is connected as a voltage follower.
-Current into the adjustment pin is small.
Coming from this the resistor from the output to the adjustment
pin will statically work as a current source.
I will try the dynamics in brain.
-Load current jumps to higher value:
==> As the regulation has a limited speed, the 1.2V will
decrease. May be to 1.19V (depending on load and open loop characteristic of the internal OP amp. The current through the
resistor from the output to the adj-pin will also decrease.
If there is no cap across the resistor to ground, then the current
in this resistor will decrease in the same proportion. And the voltage
between the adj-pin to ground will also decrease.
==> 14.875 V output until the regulation can correct the thing.
Now let's add a capacitor parallel to the resistor towards ground.
Let's think it's a big cap.
OK, the jump in the output current will decrease the 1.2V to 1.19V
again. The current in the resistor between output and adj will decrease in the same way as above. But the voltage from
the adj-pin towards ground will not drop much.
==> about 14.99V output until the regulation corrects.
In reality you only need to chose a cap which makes the
moving of the adj-pin slow compared to the regulation
of the LM317.
Do I think to simple?
For input voltage regulation I do not have a proper brain model....
I just think: It is good
Sorry.
Cheers
Markus
I am still sticking to my simplified brain-model.
Please do not misunderstand me wrong, I love PSpice.
But normally I make up my own brain model first and then
check the opinion of Spice....
Happines comes when both match somehow AND
reality proves it in the end. (Sometimes reality does
not care about my thoughts...
)My simple model brain model of the L317 is the following:
-Floating regulator
- It regulates the output 1.2V above the adjustment pin,
no matter on which potential both are.
-Internal structure can be modeled by a floating
band gap reference (or any other precise reference) followed
by an power OP amp (unfortunately without much sinking capability) which is connected as a voltage follower.
-Current into the adjustment pin is small.
Coming from this the resistor from the output to the adjustment
pin will statically work as a current source.
I will try the dynamics in brain.
-Load current jumps to higher value:
==> As the regulation has a limited speed, the 1.2V will
decrease. May be to 1.19V (depending on load and open loop characteristic of the internal OP amp. The current through the
resistor from the output to the adj-pin will also decrease.
If there is no cap across the resistor to ground, then the current
in this resistor will decrease in the same proportion. And the voltage
between the adj-pin to ground will also decrease.
==> 14.875 V output until the regulation can correct the thing.
Now let's add a capacitor parallel to the resistor towards ground.
Let's think it's a big cap.
OK, the jump in the output current will decrease the 1.2V to 1.19V
again. The current in the resistor between output and adj will decrease in the same way as above. But the voltage from
the adj-pin towards ground will not drop much.
==> about 14.99V output until the regulation corrects.
In reality you only need to chose a cap which makes the
moving of the adj-pin slow compared to the regulation
of the LM317.
Do I think to simple?
For input voltage regulation I do not have a proper brain model....
I just think: It is good
Sorry.
Cheers
Markus
Are You Experienced? ;-)
Hi ,
After 100 confusing posts may I quote from Horowitz?:
These wonderful ICs are typified by the classic LM317 from National. This regulator has no ground terminal; instead it adjusts Vout to maintain a constant 1.25 volts (bandgap) from the output terminal to the "adjustment" terminal. Fig.6.29 shows the easiest way to do it. [showing R1=240 from output to adjust and R2 =pot of 5k from adjust to ground/EK]
The regulator puts 1.25 volts across R1, so 5mA flows through it. The adjustment terminal draws very little current (50-100µA), so the outputvoltage is just
Vout =1.25(1+R2/R1)volts
In this case the output voltage is adjustable from 1.25 volts to 25 volts.
And one page further:
Improving ripple rejection
The circuit of Figure 6.29 is the standard 3-terminal regulator, and it works fine. However the addition of a 10µF bypass capacitor from the (ADJ) terminal to ground (Fig 6.31) improves the ripple and (spike) rejection by about 15dB (factor of 5 in voltage). For example, the LM317 ripple rejection factor goes from 65dB to 80dB (the latter is 0.1mV output ripple when supplied with a 1V input ripple, a typical value).
I addition I may add that no one less than John Curl provided a circuit for the powersupply of the JC2 or ML-1 Mark Levinson preamplifier with 100µF at the adjustment terminal with LM317/LM337. This circuit, which I also builded sounded better than the extremely complicated (and expensive) PLS-150 supply made by ML for the ML-1!
Now what is all the fuss about?
Hi ,
After 100 confusing posts may I quote from Horowitz?:
These wonderful ICs are typified by the classic LM317 from National. This regulator has no ground terminal; instead it adjusts Vout to maintain a constant 1.25 volts (bandgap) from the output terminal to the "adjustment" terminal. Fig.6.29 shows the easiest way to do it. [showing R1=240 from output to adjust and R2 =pot of 5k from adjust to ground/EK]
The regulator puts 1.25 volts across R1, so 5mA flows through it. The adjustment terminal draws very little current (50-100µA), so the outputvoltage is just
Vout =1.25(1+R2/R1)volts
In this case the output voltage is adjustable from 1.25 volts to 25 volts.
And one page further:
Improving ripple rejection
The circuit of Figure 6.29 is the standard 3-terminal regulator, and it works fine. However the addition of a 10µF bypass capacitor from the (ADJ) terminal to ground (Fig 6.31) improves the ripple and (spike) rejection by about 15dB (factor of 5 in voltage). For example, the LM317 ripple rejection factor goes from 65dB to 80dB (the latter is 0.1mV output ripple when supplied with a 1V input ripple, a typical value).
I addition I may add that no one less than John Curl provided a circuit for the powersupply of the JC2 or ML-1 Mark Levinson preamplifier with 100µF at the adjustment terminal with LM317/LM337. This circuit, which I also builded sounded better than the extremely complicated (and expensive) PLS-150 supply made by ML for the ML-1!
Now what is all the fuss about?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.