[cetoole]

Quote:

Originally Posted by ozshadow 4.7k R4 resistor gives this a gain of about 5.7 correct ?

Depends on the value of R3, but if R3 is 1k, then yes.

 

[muckshot]

Quote:

Originally Posted by Secret Squirrel Just Finished the "Normal" Configuration of the PPAS. It sounds nice.

Wow, that's fantastic casework! How do these compare to the PINT in a similar config?

 

[Secret Squirrel]

I have not made or listened to a “Pint”. I have made several Mints, PPA, Pimeta, CMOY, Millet Hybrid, and various iterations of Sijosae’s tube designs. I like the PPAS as an overall portable amp.

 

[pddjsteve]

Ok, so I'm about to start putting together my 2nd one. I'm going to continue using the LMH6321 buffers. I've got 2x470uF panasonic FM rail capacitors. I've got an unregulated 24v 100mA wall-wart and as soon as they arrive, a battery pack made from 12xAAA solder tab nimh (700mAh). I'm probably going to put together a partial TREAD, I've already got a pint-style charging circuit. I listen through ER4P and a homemade 75ohm P->S adapter.

I haven't decided on opamps. My first one used AD8620/8610. I really like the way it sounds, except for a little bit of harshness on the highs. I'd like to try something else, though.

I know that someone has tried ad8066 and someone else tried opa2107. What are your impressions of either of them? For ground, would an opa227 work in combination with either of those two?

 

[quicksilver96]

Finally built one of my PPAS boards and cased it up in an aluminum stomp box, pictures are in the big post pics thread. Here is my configuration for anyone curious:

R1/R1G/R4G - 4.42kΩ
R2 - 1MΩ
R3 - 402Ω
R4R/L - 2.21kΩ
R5 - 4.02kΩ
R6 - 1MΩ
(I believe I managed to find .1% vishay in all these values at mouser)

Gives the following specs-
Outer Loop Gain - 6.47
Inner Loop Gain - 249.76
Total Gain - 6.322

AD8620/10 GBP - 25Mhz
Amp Bandwidth - 100kHz

Full volume offset is around .1mV on the left .6mV on the right. The amp consumes right around 16.5 mA of current with the buffer bandwidth jumper position open.

Build impression - Build is fairly simple for someone with building experience. Not the ideal intro to SMD with 0805 resistors but for someone who has worked with 1206, there shouldn't be much trouble. I don't know if there is a guide to building this, but I found it easier to solder on resistors, diodes, and jfets before the opamps and the smd capacitors. The soic parts are actually some of the taller components on this board and reaching over a 0805 or even 1206 resistor to get to the soic leads seemed less of a pain in my ass than having the ICs in the way of the resistor pads. Of course the through hole crap goes on last.

On a side note, I squeezed the 16V 180uF (6.3x15) Nichi UPWs in the spot for the 10v 100uF caps (5x11) so I could use a 24v steps to run the amp while charging if needed. This means that the 25v 120uF UPW will also fit and give an extra measure of safety should the rail splitter fail and hit those caps with full rail.

Listening Impression - Well its AD86x0 through BB buf634s. If you've ever heard a Pimeta with the same setup, expect much the same. The buffers smooth out the 86x0 pleasantly. I could not tell any difference in sound with the buffers running in high bandwidth so I opened the jumper back up again. I ran it A/B next to my PINT and it would be cake to pick them out blind. The PINT has a fuller, though less precise, bass and a more forward midrange, the overall sound seems slightly blended together. The PPAS has a slightly thinner sound overall with much better instrument seperation, more sparkle on the highs, and a very precise, well defined bass through midrange. I am quite impressed by the last. The smearing/blending on the PINT may be caused by the input caps where the PPAS really doesn't have any caps directly in the signal path. Seeing as how these are portable amps, I would have to say the dealbreaker for the PINT is current consumption as it uses nearly 3x the current of the PPAS.

Impressions are with K701s and my modded av710. Tomorrow I think I am going to pick up a pair of UE IEMs to give a try seeing as I can buy them locally and shouldn't have problems returning them if I hate them. If anyone can point me in the direction of some IEMs sub $300 that have a sound signature close to K701s...

 

[ozshadow]

Has anyone built using a 10k Alps pot and changed R2 to just 100k ? I do not know how noticeable it would be, but techinically it should reduce noise even if a minute amount. A 1M resistor would be advisable for 100k pot.

 

[pddjsteve]

Quote:

Originally Posted by ozshadow Has anyone built using a 10k Alps pot and changed R2 to just 100k ? I do not know how noticeable it would be, but techinically it should reduce noise even if a minute amount. A 1M resistor would be advisable for 100k pot.

I used 475k instead of 1M. Dunno what effect it might have had.

 

[ozshadow]

what has your total cost so far ? Mines gonna be way way more than $50. Like 150+

 

[dhp]

Quote:

Originally Posted by ozshadow what has your total cost so far ? Mines gonna be way way more than $50. Like 150+

 

[pddjsteve]

Quote:

Originally Posted by ozshadow what has your total cost so far ? Mines gonna be way way more than $50. Like 150+

I think I'll be around $100-120 for two of them by the time I order BUF634 for my 2nd.

 

[ozshadow]

I didn't see any decoupling caps. Do the buffers act in a similar manner to remove DC offset ? I am curious about whether I need to add caps in the signal path of my Alien DAC.

 

[cetoole]

Quote:

Originally Posted by ozshadow I didn't see any decoupling caps. Do the buffers act in a similar manner to remove DC offset ? I am curious about whether I need to add caps in the signal path of my Alien DAC.

I think you are confusing some terms here. I suspect you mean coupling caps, which are used to remove DC from the signal, and there are none of these, or any other method of removing DC offset in this amplifier. The buffers certainly dont help lower DC offset, and in fact, it is only the fact that they are being run closed loop that prevents them from exhibiting a small offset of their own. That said, you dont need to add coupling caps to your Alien DAC, but you must do something to block the DC it puts out, or fry your headphones, which is most easily solved by coupling caps, though could also be solved by playing games with a virtual ground on the DAC floating midway between ground and supply, equal to the voltage of the offset. This cant be implimented on the existing Alien DAC pcb, so you would have to either hack the board, do it offboard, or design a new pcb with that in mind. Or just add a pair of coupling caps, and dont worry too much about it.

 

[ozshadow]

This is for a PPAS in a tiny hammond case with charging circuit running right at 27 volts (want the tread to be able to power an AD8620/10 version too) I have several caps listed due to needing to make sure one set will squeeze in the case.

Think this is all my parts for one of them:

OPA2107
OPA627 (seriously considering an AD8610 as well)
Tread Kit
30v AC/DC plug
2 Accupower 9v 270mah
Buf634u
Silver Knob JD-50-1-6MM
Alps RK097 pot

Mouser:

71-CRCW0805-4.32K
Vishay/Dale 0805 Thick Film Chip Resistors
1/8WATT 4.32KOHMS 1%


71-CRCW1206-1.0M
Vishay/Dale 1206 Thick Film Chip Resistors
1/4watt 1.0Mohms 1%


71-CRCW0805-1.0K
Vishay/Dale 0805 Thick Film Chip Resistors
1/8WATT 1.0KOHMS 1%

71-CRCW0805-10K-E3
Vishay/Dale 0805 Thick Film Chip Resistors
1/8watt 10Kohms 1%

71-CRCW1206-100K
Vishay/Dale 1206 Thick Film Chip Resistors
1/4watt 100Kohms 1%

581-TAJA105K020
AVX SMD Tantalum Capacitors
20V 1.0UF 10%

581-08055A100J
AVX 0805 SMD Ceramic Capacitors
50V 10pF 5% COG –

781-SST502-E3
Vishay/Siliconix Diodes
SOT-23 0.43mA Current Reg

512-BSR57
SOT-23-3 N-Ch Transistor Lo Freq/Lo Noise –

161-7300-EX
Kobiconn Phone Jacks
3.5MM STEREO

604-WP7104QBC/D
Kingbright LED Blue
T1 BLUE WATER CLEAR

546-1455C1201BKHammond Instrument Enclosures
4.72 X 2.125 X 0.91 Black Anodized –

512-LM317T
TO-220 TO-220 1.5A ADJ VREG –

512-1N4001
DO-41 Vr/50V Io/1A T/R

163-4303-EX
Kobiconn DC Power Jacks
PANEL MOUNT 2.5mm –

71-CRCW0805-3.32K-E3
Vishay/Dale 0805 Thick Film Chip Resistors
1/8watt 3.32Kohms 1%

647-UPW1C181MED
Nichicon Low Impedance Radial Aluminum Electrolytic Capacitors - 105Degree
16V 180uF 6.3x15 –

71-CRCW0805-4.7K-E3
Vishay/Dale 0805 Thick Film Chip Resistors
1/8watt 4.7Kohms 1% -

647-UPW1V331MPD6
Nichicon Low Impedance Radial Aluminum Electrolytic Capacitors - 105Degree
35V 330uF 8x20

647-UPW1V331MPD
Nichicon Low Impedance Radial Aluminum Electrolytic Capacitors - 105Degree
35V 330uF 10x16

647-UPW1V221MPD6
Nichicon Low Impedance Radial Aluminum Electrolytic Capacitors - 105Degree
35V 220uF 8x15

 

[quicksilver96]

Looking at your list, I would say you are running this:

R1 = 4.32k
R2 = 100k
R3 = 1k
R4 = 10k
R5 = 3.32k
R6 = 1M

OLG = 11
ILG = 302.2
Total Gain = 10.556

OPA2107 GBP = 4.5MHz
Amplifier Bandwidth = 14.89 kHz

Is there any reason that you don't optimize the multiloop values? With these resistances, your amplifier is going to suffer from rolled off high frequencies if you use a relatively slow opamp like the OPA2107. These values are from the Pimeta schematic and intended for use with an AD8620/8610 family of chips with a GBP of 25MHz where the OPA2107 only operates at 4.5Mhz. Generally, you want to adjust these values for a amplifier bandwidth at least twice the audible frequency band of 20kHz -- so minimum 40kHz bandwidth, though most shoot for 70-80kHz as this bandwidth limit is not a brick wall and opamps don't necessarily function at thier exact GBP rating.

Second, is there any reason that you chose such a high gain (10.5)? Weak source? Voltage loving phones? If you really want a gain this high, let me know and I will suggest some resistor values that will be more bandwidth friendly.

Another thing to look at is your chosen voltage, AD8620/8610 are specified a maximum +/- 13v supply (26v). The ABSOLUTE maximum is 27.3v, normal operation at 27v is not acceptable for this chip. If you go with these chips rather than the OPA2107/627 combo you have chosen, you will need to reduce the supply voltage. There is no performance benefit to outweigh the damage potential.

Last, the 10mm caps will not fit with the 6mm caps in the normal position. You will have to bend the leads or topple the caps to use this configuration. With 8mm and 6mm, the capacitors will be pushing against each other and require minor lead adjustment. If you are planning on this, then you know the risks. Not that capacitors are very expensive anyway.

 

[ozshadow]

I am using the 4.7k resistor to set a gain of just under 6

I do not understand the idea of optimizing the resistor values. What do you suggest and what differences would it make ?

I will be running the tread at about 26.6 which by the time it comes out of the charging circuit, it should be about 26.0 flat.