[nightanole]

Quote:

Originally Posted by nightanole /img/forum/go_quote.gif Well i got everything ordered. I felt dirty ordering the capacitor bank of PW series caps with .23 resistance, at least the ripple of 430 will help me sleep at night. I looked for like 2 hours and couldnt find a better cap, everything needed a 8mm can. The only thing i found that might work was using 2 1800uf nichicon UHZ caps 16v. then it would be .0065 resistance and 4140. They are 10x25mm so you would have to lay them on their sides, but it would double the current compaired to the stock bom. UHZ1C182MPM Nichicon Aluminum Electrolytic Capacitors - Leaded

Im weak. I added the caps to my backordered set of opamps that are due in october.

I should go from a stock

From the looks of the board, I just need to add a cap to the c9-12 bank and one to the c15-20 bank.

the Stock BOM bank is 19.17mohm , 5160ma ripple, 2160uf
My bank should be 3.25mohm, 8280ma ripple, 3600uf.

I think i might of even bested the 10v panasonic fm bank.

10.83mohm , 5460ma ripple, 2640uf

 

[joneeboi]

Interesting implementation, nightanole. Maybe I'll give it a whirl for the production version if it fits okay on the board.

 

[cobaltmute]

Quote:

Originally Posted by nightanole /img/forum/go_quote.gif the Stock BOM bank is 19.17mohm , 5160ma ripple, 2160uf My bank should be 3.25mohm, 8280ma ripple, 3600uf. I think i might of even bested the 10v panasonic fm bank. 10.83mohm , 5460ma ripple, 2640uf

Total Capacitance of the bank is not that high.

In the Carrie, we've go 6xcap from V+ to Vgnd and 6xcap from Vgnd to V+

This puts the two banks in series with each other. For my 220uF FM caps that makes a bank of 1320uF and 1320uF in series. Or 660uF in effective total for the whole bank (if I did the math right).

 

[nightanole]

Isnt it 6 caps from +5v to neutral ground and 6 caps from v ground to -5v. 12 caps in parallel across 10v rails? Wouldnt that be considered parrellel if it werent for a neutral ground center tap? Or should i not compare this to a transformer center tap?

 

[joneeboi]

I think the opamp sees both banks in parallel. Both maths are right, you just need to state how you're comparing the rail banks. If you're talking series from V+ to V-, you take half the capacitance. If you are talking about the banks in parallel, you add them. But if we're speaking relative the opamp, it sees a parallel connection of banks from which it can draw energy. Again, someone correct me if I'm wrong.

 

[cobaltmute]

The opamps have no sense of ground. They pull from V+ and V- and the bank is in series between the two rails with 6 caps each.

"Capacitors on the Output of a Virtual Ground Driver" on Tangent's Virtual ground page explains more, especially the last paragraph of that section.

 

[joneeboi]

Riiight. Hadn't thought of that. For that matter, maybe I should throw some of those caps across the rails. There's already CF2 across the rails, but more capacitance would be helpful for the instantaneous currents to the opamps. Hm...

 

[nightanole]

Well lets do a single bank of 6 for giggles.

Stock pw's:
38.3mohm 2580ma 1080uf


panasonic FM:

21.6mohm 2730ma 1320uf

single UHZ:

6.50mohm 4140ma 1800uf

 

[cobaltmute]

Before you get too capacitance happy, start thinking about in-rush current. I just measured my Carrie at pulling 1A at startup. Windows 7 shuts the USB port down...

 

[nightanole]

hmm well you do have the 2 watt converter version. The 1 watt peaks at 500ma. I ordered both the stock caps and the 2 that i found.

At some point someone is gonna say bypass the usb converter and use a 12v tread

 

[DKJones96]

Wait, 1A? The regulator shouldn't be able to flow that at all since it's rated at 150ma. Fortunately, there is 2.5 watts available through the USB and we are only using 1w of it with the DCP, even running double capacity it should still be OK. Close, but nothing obscene like 1A.

Maximum startup current from this amp plus whatever capacitors used in your dac charging are causing issue?

 

[cobaltmute]

Quote:

Originally Posted by nightanole /img/forum/go_quote.gif hmm well you do have the 2 watt converter version. The 1 watt peaks at 500ma. I ordered both the stock caps and the 2 that i found.

Good point at the 2W convertor. Though it in theory should output 0.222A. Apparently in that first split second it takes everything a second to really settle down. If I had used the switched pot, I could plug in the USB and then turn on the rest of the amp.

Try to figure out how to current limit the startup as once it is active, the Carrie is pulling 70mA in my configuration.

 

[cobaltmute]

Quote:

Originally Posted by DKJones96 /img/forum/go_quote.gif Wait, 1A? The regulator shouldn't be able to flow that at all since it's rated at 150ma. Fortunately, there is 2.5 watts available through the USB and we are only using 1w of it with the DCP, even running double capacity it should still be OK. Close, but nothing obscene like 1A. Maximum startup current from this amp plus whatever capacitors used in your dac charging are causing issue?

Yup, 1A measured by,

- disconnecting the Carrie from the DAC
- attaching the bench supply to the Carrie (5V, 4A)
- placing my Fluke in between to measure current and turning on peak hold.
- turning on the power to the bench supply (only way I can switch it as my Carrie is unswitched)

I don't get how it is pulling that much.

 

[nightanole]

Ya know i went over the cap bank again. In the mini 3 the cap bank is across the rails, with just a 100uf at the rail splitter. This amp is just the opposite.

Im wondering if i should just shove ambs 100uf caps in c9,c18 and pop my big 2 caps across the rails.

 

[DKJones96]

What regulator are you using?