DCAudioDIY.com

DC Area Audio DIYer's Community
It is currently December 5th, 2019, 7:12 pm

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 44 posts ]  Go to page 1, 2, 3, 4, 5  Next
Author Message
PostPosted: February 26th, 2019, 8:45 pm 
Offline
Site Admin
User avatar

Joined: February 28th, 2013, 10:38 am
Posts: 1169
Whipped up this PS design today, probably order the parts next week.... Thoughts?

Roscoe


Attachments:
op-amp phono ps-1.jpeg
op-amp phono ps-1.jpeg [ 86.17 KiB | Viewed 2229 times ]

_________________
I can explain it to you, but I can’t understand it for you.
Top
 Profile  
 
PostPosted: February 26th, 2019, 9:56 pm 
Offline
User avatar

Joined: February 28th, 2013, 3:31 pm
Posts: 1184
Am I reading this correctly? Are those 90 farad caps in series? Regardless of the value you might want to put balancing resistors across them to handle any leakage imbalances which would affect the voltage across each cap.


Top
 Profile  
 
PostPosted: February 26th, 2019, 10:06 pm 
Offline
Site Admin
User avatar

Joined: February 28th, 2013, 10:38 am
Posts: 1169
tomp wrote:
Am I reading this correctly? Are those 90 farad caps in series? Regardless of the value you might want to put balancing resistors across them to handle any leakage imbalances which would affect the voltage across each cap.


You read it right... The docs for the caps say balancing isn't necessary, the 8.4v rating is actually 6*1.4 in series anyway...

_________________
I can explain it to you, but I can’t understand it for you.


Top
 Profile  
 
PostPosted: February 26th, 2019, 10:20 pm 
Offline
User avatar

Joined: July 24th, 2015, 4:17 pm
Posts: 951
Location: Parkville, Maryland
Roscoe Primrose wrote:
Whipped up this PS design today, probably order the parts next week.... Thoughts?

Roscoe


I see where you are going with 22.5-Farads for each half of the +/- power supply -- but -- and it's a big BUT -- can those caps handle the ripple current?

By design they are intended for "battery" back-up for volatile memory not power-supply filtering. Ripple current beyond what the caps are rated for (if they are rated for ripple at all) can cause dielectric failure mainly from internal heating.

And yes -- Tom is correct -- you need balancing resistors like about 100Kohm across each capacitor. To quote Illinois Capacitor:

"Ideally when capacitors are placed in series the voltage across each capacitor should be distributed evenly across the capacitors. This does not always occur as intended due to the variations in the leakage currents of the capacitors. The variations in leakage current can lead to some of the capacitors having voltages across them that could exceed their voltage ratings. Exceeding the voltage ratings of a capacitor will compromise the reliability of the capacitor and could have other severe and dangerous consequences. It is therefore important that this situation be avoided in any application."

_________________
Walt


Top
 Profile  
 
PostPosted: February 26th, 2019, 10:30 pm 
Offline
Site Admin
User avatar

Joined: February 28th, 2013, 10:38 am
Posts: 1169
The charger will be regulated, so the ripple should be in the mV region... From the first page of the Vishay datasheet: "Cell balancing is not necessary" https://www.mouser.com/datasheet/2/427/196hvc-463326.pdf If you drill down into some of the supercap/ultracap datasheets, that's because the cells already have internal parallel resistors. However, with the larger caps than I was planning (there's no affordable caps between 15F & 90F, and the 90F caps are <2x the cost of the 15F caps, go figure) I can afford a little extra current in balancing resistors of my own... If I did the back-of-the-napkin math right, I ought to get >4hrs run time before I get close to regulator drop-out. If you top it off while you're flipping the record, it'll probably never get low enough to be an issue....

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.


Top
 Profile  
 
PostPosted: February 26th, 2019, 11:39 pm 
Offline
User avatar

Joined: December 14th, 2013, 2:19 pm
Posts: 743
Well Roscoe,I don't want to try and dissuade you from doing over the top dumb shit, I have a huge amount of personal experience in that field!

1) Hope that operate/charge switch is an oil bath contactor, because you are going to arc the hell out of the contacts every recharge cycle! Adding a relatively large shunt resistor across each pair of contacts, AND using a 4-pole high current switch, configured as two parallel sets of contacts should help.

2) Not sure why you chose the LMXX series of regs, or that configuration. May I offer that using AT LEAST a LT1086 type should give a quieter regulator.

3) As to your regulator configuration, may I suggest applying Dave Berning's rule and decouple all loads from a) the regulator and, b) from each other. To that end, why not just use one regulator per channel, per rail, then RC decoupling stages for each op amp? Use nice QUIET wirewound resistors (the more inductive, the better) and low ESR caps after the regs, 220uF per rail/stage. This should shunt out regulator noise. In a power supply application, as Roscoe knows very well, series inductance can be your friend, attenuating HF.

And noise is an extremely important aspect of any stage being asked to amplify signals smaller than 0.1mV 40-60dB.
Hence my surprise at your using undecoupled LM7818 regs.

Just my 0.02

Stuart


Last edited by Stuart Polansky on February 27th, 2019, 8:41 am, edited 1 time in total.

Top
 Profile  
 
PostPosted: February 27th, 2019, 8:10 am 
Offline
User avatar

Joined: July 24th, 2015, 4:17 pm
Posts: 951
Location: Parkville, Maryland
Stuart Polansky wrote:
Use nice QUIET wirewound resistors (the more inductive, the better) and low ESR caps after the regs, 220uF per rail/stage. This should shunt out regulator noise.

And noise is an extremely important aspect of any stage being asked to amplify signals smaller than 0.1mV 40-60dB.
Hence my surprise at your using undecoupled LM7818 regs.

Just my 0.02

Stuart


Just a note about the frowned-upon wire-wound resistor. You want transparency and quiet? Wire-wound resistors are the way to go. You can get the inductive types (basic) or the non-inductive types (better). And if you want to fly first class (signal path) the bobbin wire-wound resistors will surpass the very best that Vishay bulk-film resistors have to offer.

_________________
Walt


Top
 Profile  
 
PostPosted: February 27th, 2019, 2:03 pm 
Offline

Joined: February 24th, 2018, 6:16 pm
Posts: 19
Yikes! That would some filtering!

I've thought about something like this only using rechargeable nimh batteries instead of those enormous capacitors. No 60 cycle lines garbage if the thing is essentially battery operated!

I think I'd use one regulator and some decoupling resistor/capacitors. For one thing it would be simpler and for another the current draw of the regulators is probably going to be greater than that of the op amps.

And, as others pointed out there are significantly quieter regulators!

Mark


Top
 Profile  
 
PostPosted: February 27th, 2019, 2:12 pm 
Offline
User avatar

Joined: February 28th, 2013, 3:31 pm
Posts: 1184
The choice is about location. Trying to get rid of line noise before the regulators is one option but then leads to having to deal with regulator and other noise after the regulators. Or you can treat the input supply as a simple source and do all the noise filtration after the regulators and right at the Op Amps. The closer you get to the amps the more rejection of all noise you get including that from current draw of the individual amps.


Top
 Profile  
 
PostPosted: February 28th, 2019, 9:15 pm 
Offline
Site Admin
User avatar

Joined: February 28th, 2013, 10:38 am
Posts: 1169
OK, I'm gonna reply out of context to several of the comments...

1. They're not filters, they're energy storage. I didn't use NiMH/LiPo etc because the charging requirements are much more of a pain in the butt, and the supercaps should be good for >500K charge cycles.
2. As for the current surge when charging, the power supply is going to have a 15ohm output impedance, so the surge will be quite small.
3. OK, you guys have convinced me, I'll use TL431s with pass transistors, one per rail, and individual LC (yeah Stuart, wire wound resistors with LOTS of inductance) filter between the regulators and each op-amp.
4. Line noise? What line noise? The charger will be switched out when the phono stage is operating, and only connected when recharging.

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 44 posts ]  Go to page 1, 2, 3, 4, 5  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group