Re: Modifications
Is there a necessity to placing the two zeners in parallel? If the diodes aren't matched one will tend to hog current.
It may be gilding the lilly, but here's a LM317 regulator cleanup circuit where the noise is measured in nanovolts, the performance only limited by the opamp used and various resistances:
DcibeL said:
Is there a necessity to placing the two zeners in parallel? If the diodes aren't matched one will tend to hog current.
It may be gilding the lilly, but here's a LM317 regulator cleanup circuit where the noise is measured in nanovolts, the performance only limited by the opamp used and various resistances:
An externally hosted image should be here but it was not working when we last tested it.
Member
Joined 2003
The Zeners are there to provide a very low dynamic impedance at the adjust pin, so two in parallel will lower this impedance even more. It has nothing to do with current flowing through them, so if one hogs a bit there's no worry. I don't really even need the 10uF cap in there when using zeners instead of a resistor, but it's not going to hurt anything so why not.
See here for more information on using the LM317/337. I was thinking of building a super regulator, but I do want it to fit easily in the current regulator location, and it would just have too many parts to make this possible (plus add a fair bit to the cost).
I built the regulators last night, I just have to cut them out of the veroboard and test them now.
See here for more information on using the LM317/337. I was thinking of building a super regulator, but I do want it to fit easily in the current regulator location, and it would just have too many parts to make this possible (plus add a fair bit to the cost).
I built the regulators last night, I just have to cut them out of the veroboard and test them now.
Banned
Joined 2002
Member
Joined 2003
Banned
Joined 2002
If you really want to improve the reference use something like an embedded reference and some temperature compensation. 2 zeners isn't going to do much.DcibeL said:
Bob Pease has a nice piece on the why's of bypassing the LM317 adjust pin on the National Semi website -- here's a demonstration of the effect using a 5223 storage scope and 5a22n differential amplifier:
No bypass:
An externally hosted image should be here but it was not working when we last tested it.
220uF bypass:
An externally hosted image should be here but it was not working when we last tested it.
Member
Joined 2003
A couple of zeners will do a better job at bypassing the adjust pin than a resistor and a cap will. I found some 47uF low ESR caps left over from other builds that I will use in place of the 10uF cap.
I wanted to build something small that could easily be substituted for the 78/79xx regulators that was better, and I believe this circuit will fit nicely. I'm not going to change it now, since the regulators are nearly complete.
I wanted to build something small that could easily be substituted for the 78/79xx regulators that was better, and I believe this circuit will fit nicely. I'm not going to change it now, since the regulators are nearly complete.
Member
Joined 2003
DcibeL said:
The impedance of the regulator circuit is entirely dominated by C1/C2. The junction capacitance of a 1W zener is pretty small (a few pf).
You might get a hair lower noise since the zener noise of the two diodes are uncorrelated.
Now you might want to sweep that circuit over temperature and tell us how the two parallel zeners perform.
Hi again,
I made a USB NOS DAC plus headamp for my Dad who is a dentist about to retire: he began to love navigating through Youtube and listening some opera singer from the past...just as his son!
USB receiver comes from Doede, NOS DAC is self made "á la Doede" and headamp is the famous Kookaburra with OPA2227 opamps.
I used a front plate which belonged to his ancient dental machine, Japanese, very well made...
The box was difficult to make even with wood parts...two stores.
The sound is warm and inviting, not the best in detail but one that lets get listened to very easily, and the bass is terribly good.
Cheers,
M
I made a USB NOS DAC plus headamp for my Dad who is a dentist about to retire: he began to love navigating through Youtube and listening some opera singer from the past...just as his son!
USB receiver comes from Doede, NOS DAC is self made "á la Doede" and headamp is the famous Kookaburra with OPA2227 opamps.
An externally hosted image should be here but it was not working when we last tested it.
I used a front plate which belonged to his ancient dental machine, Japanese, very well made...
The box was difficult to make even with wood parts...two stores.
The sound is warm and inviting, not the best in detail but one that lets get listened to very easily, and the bass is terribly good.
Cheers,
M
I was inspired by the Kookaburra design to use the same PIC and code to control two PGA2310's. The difference is that I have the digital logic going straight to two PGA2310's. I didn't use pull-up resistors (47k) or series resistors (10k) as shown on TI PGA2310-EVM (eval board.) I don't know if these are required but I am not getting the VC's to control signal level at all with a perfectly good 0-5V input to PIC. I also used a potentiometer and ADC inside PIC as Russ & Brian have done. Are those resistors most likely the missing link?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.