Re: Adjustable etc
Exactly and also the way it is explained in most documents describing the pass transistor. As for the K version I think the reason as I mentioned simply is because it is one of the versions in this datasheet. Nothing to complicated, it just happens to be the K version drawn. Simple enough.
Thanks for filling in Jan.
janneman said:
Exactly and also the way it is explained in most documents describing the pass transistor. As for the K version I think the reason as I mentioned simply is because it is one of the versions in this datasheet. Nothing to complicated, it just happens to be the K version drawn. Simple enough.
Thanks for filling in Jan.
Roady? 
😀
Thanks for the tutorial, jan!
I have simulated (in Spice) a dual-tracking regulator, which can current-limit only the positive rail for now. I will post it here when I can get the current limit to also work for both rails.
😀 Thanks for the tutorial, jan!
I have simulated (in Spice) a dual-tracking regulator, which can current-limit only the positive rail for now. I will post it here when I can get the current limit to also work for both rails.
If this is a lab supply, here are some points to consider:
1) by biasing the ADJ pin to -1.25 volts you can get down to "0".
2) these LM317 circuits were OK when folks used analog meters with 1% or 2% acuracy. Now, with a 10 bit ADC you should be able to get close to 0.1%, with 12 bits you will get better than 0.1% etc. If the supply is going to drive some audio circuitry, say a phono-preamp, or to breadboard and test a preamp, you should really do better than 0.1%. Consider using the noise-cleanup circuitry that Wenzel describes on their site:<p>http://www.wenzel.com/documents/finesse.html
1) by biasing the ADJ pin to -1.25 volts you can get down to "0".
2) these LM317 circuits were OK when folks used analog meters with 1% or 2% acuracy. Now, with a 10 bit ADC you should be able to get close to 0.1%, with 12 bits you will get better than 0.1% etc. If the supply is going to drive some audio circuitry, say a phono-preamp, or to breadboard and test a preamp, you should really do better than 0.1%. Consider using the noise-cleanup circuitry that Wenzel describes on their site:<p>http://www.wenzel.com/documents/finesse.html
jackinnj , a properly designed amp (phono or otherwise) should not affected by variations in the supply voltage, of 1% or even more. This is assuming you mean LM317s suffer from 1% DC inaccuracy.
Also, why would you want to drive an amp with a variable supply anyway? (except during experimenting, but then you should have a proper design, supply and all, before starting to test it for noise).
The ADCs used in audio circuits are usually driven by an opamp at the input, which will have a high PSRR. PSRR at the top end of the audio band for most opamps is of the order of 30dB, worst case. This will drive the already low power supply noise to inaudible levels, if not, below the noisefloor.
Also, why would you want to drive an amp with a variable supply anyway? (except during experimenting, but then you should have a proper design, supply and all, before starting to test it for noise).
The ADCs used in audio circuits are usually driven by an opamp at the input, which will have a high PSRR. PSRR at the top end of the audio band for most opamps is of the order of 30dB, worst case. This will drive the already low power supply noise to inaudible levels, if not, below the noisefloor.
Didn't I say "lab supply"?
At any rate, take your oscilloscope probe and noodle around. You shouldn't be surprised at the junctions where noise crops up.
At any rate, take your oscilloscope probe and noodle around. You shouldn't be surprised at the junctions where noise crops up.
Yes, you were right UrSv
I did not know about the Current Limiting thing.
My logic said me, that if you have 2 TO3 devices
it is bad design to let one of them carry all the burdon
in this case power/heat.
If you have two TO3 parallell you divide the power
so that both devices operate safe, clear within
their limits.
But if Current Limit is only in one of the rails
of course you direct as much of the current
as possible into that one.
That circuit was a sample. But I guess any sensible
designer wouldn't use a TO3 just for controlling purposes.
As TO220 and TO92 has better performances
when it comes to such tasks.
They are mostly both faster and have other prestanda
that TO3 don't have.
So, Excuse me UrSv for not checking up facts better
I am not "a bad loser".
/halojoy
in the north of our beloved country
fredagen den 17:e januari 2003
I did not know about the Current Limiting thing.
My logic said me, that if you have 2 TO3 devices
it is bad design to let one of them carry all the burdon
in this case power/heat.
If you have two TO3 parallell you divide the power
so that both devices operate safe, clear within
their limits.
But if Current Limit is only in one of the rails
of course you direct as much of the current
as possible into that one.
That circuit was a sample. But I guess any sensible
designer wouldn't use a TO3 just for controlling purposes.
As TO220 and TO92 has better performances
when it comes to such tasks.
They are mostly both faster and have other prestanda
that TO3 don't have.
So, Excuse me UrSv for not checking up facts better
I am not "a bad loser".
/halojoy
in the north of our beloved country
fredagen den 17:e januari 2003
Attachments
jackinnj, I guess you did.
<- for not reading posts right.
halojoy, I doubt the case type could make a difference to a device's "speed" (I assume you mean risetime or some other measureable parameter), unless we're talking RF.
<- for not reading posts right.halojoy, I doubt the case type could make a difference to a device's "speed" (I assume you mean risetime or some other measureable parameter), unless we're talking RF.
Whatever.roadkill said:jackinnj, I guess you did.
halojoy, I doubt the case type could make a difference to a device's "speed" (I assume you mean risetime or some other measureable parameter), unless we're talking RF.
You can use as many TO3:s you like
/halojoy - is not bothered by that
- Status
- Not open for further replies.