[picklgreen]

I finally got around to finishing my PPA! I actually forgot all about it. I started it a long time ago so it was built on the V1.0A board. I guess they have since had 2 major revisions. Oh well....

So it powered up right off the bat and sounded pretty good. I prefer the sound of the OPA 627.637 mix. Just the 627 in all 3 channels was to laid back for me.

Had a bit of a problem doing the class A bias but i got help with that. Have them set to 2mA bias.

I will post some pics when i figure out how to post them.

For my next project I would like to build either the PPA V2 or an M3. Any thoughts?

 

[dhp]

anyone have 3 2N5088s and a 2N5087 transistor? They're used in the discrete buffer stages of the PPAv2. Shoot me a PM and it'll be greatly appreciated. Thanks

 

[luvdunhill]

reviving a dead thread

Anyways, can someone give me some insight in how to choose a CRD for the PPAv2? values go from 0.24mA - 4.7mA in the range given on Tangent's website. My gut is to go for a higher value, but I thought I'd ask and get the mathematical answer

 

[tangent]

Quote:

Originally Posted by luvdunhill reviving a dead thread Anyways, can someone give me some insight in how to choose a CRD for the PPAv2? values go from 0.24mA - 4.7mA in the range given on Tangent's website. My gut is to go for a higher value, but I thought I'd ask and get the mathematical answer

There are several places in the PPA you could use a CRD. Which one are you talking about?

 

[luvdunhill]

In the PPAv2 PCB position marked CRD. It's for the LED

 

[tangent]

Okay, so technically, there is indeed a mathematical answer. You would have to talk the LED's manufacturer out of its lumens vs. current graph -- I've yet to see one in a datasheet -- and then decide how many lumens is "bright enough" for you. That'll give you current.

Or you can do what the rest of the world does and fiddle with it until you find a brightness level that makes you happy. With most LEDs and most eyes, this will be somewhere around 2 mA. It may need to be as high as 5mA, and some may tolerate values under 1 mA. You'd do this with a resistor and a fixed power supply voltage, then calculate the correct CRD value from that.

If you must know the correct answer before buying parts, you're out of luck. There is no easy answer. If you had to pick a value blindly, I'd do 1 mA for a red, green (the sickly yellowish green traditional color -- not the new "true" greens) or amber LED. For the newer types -- white, blue, pure green, violet... -- I'd probably do 2 mA.

 

[ozshadow]

I have yet to find the perfect blue LED. The 1500 brightness 5mm Radio Shack ones light up half my bedroom at night with a 10k resistor - and its purple looking.


Trying this one next with 10k resistor
http://www.mouser.com/search/Product...04-WP7104QBC/D

 

[picklgreen]

I have several different blue leds in 3mm and 5mm that work great! And yes they are brighter than red ones but that is because of the spectrum of light blue falls in. But when i use resistors, it does not make them purple...they are still blue!

I have some pink ones also that look sweet.

 

[luvdunhill]

Hi all fellow PPA builders

I was wondering if using a 475k instead of 470k resistor for R6RL would make any difference other than a slight increase in gain?

Also, when connecting the BBR and R7R pads, should I run the wires on the bottom or the top of the board? Also, Tangent suggests testing the amp before adding the bass boost, but I should be able to just jumper the S2RL pads and still run the BBR->R7R wires before hand, right? In other words, I'd rather add the BBR-R7R wires, jumper the BBR->R7R, test the PPA, unjumper and test the bass boost.

Thanks!!

 

[tangent]

Quote:

Originally Posted by luvdunhill I was wondering if using a 475k instead of 470k resistor for R6RL would make any difference

Not a whit.

Quote:

should I run the wires on the bottom or the top of the board?

I'd do it on the bottom, only because there aren't any components in the way. If you had to do it on the top for some reason, there's no great problem with that.

Quote:

Tangent suggests testing the amp before adding the bass boost, but I should be able to just jumper the S2RL pads and still run the BBR->R7R wires before hand, right? In other words, I'd rather add the BBR-R7R wires, jumper the BBR->R7R, test the PPA, unjumper and test the bass boost.

I'm not understanding which end you're jumpering.

If you're soldering the boost control wires together, then no, that's not a good plan. Part of the reason I say you should test without bass boost is because the smaller the feedback loop, the fewer the chances for problems.

If you're shorting it so the wires aren't in the loop, I suppose that would be okay, but I don't know that you'd get a great amount of advantage from this. Regardless of how you do this, you're going to have to desolder something, so...

 

[luvdunhill]

hi Tangent! Basically, I was wondering if I could go ahead and run the two wires from BBR to S2R and *also* jumper S2L and S2R.. test and then remove S2L/S2R to test the bass boost.. I just would like to go ahead and run the wires now

Since you recommend putting them on the bottom of the board, I'm not sure how to do this!! Do I solder on the front side of the board? There's no way I can really do this with all the parts already on the board.. so I assume that I would just tin the wire, shove it in and flow solder around the wire on the back?

Also, I'm thinking about trying the OPA637BPG4 op amps first (I have no idea what the G4 means, as there's no difference I can see on TI's website) so wish me luck! I also have OPA134PA and OPA227PA single channel opamps on hand as well if the former turns out to be cranky!

 

[luvdunhill]

hm, quick question. What could cause a different DC bias in one channel but not the other? I have 2.2mV in the right channel and 0.5mV in the left channel. I have set R24/R34 to 44mV in all three channels and have biased the opamps with 1.0V across R10 in all three channels as well. I'm just using two 9V batteries right now to test, until my Elpac arrives, in case that is of interest.

edit: my input voltages are off my a half volt or so to my opamps, so I'm going to assume that I testing with two 9v batteries in series is the problem. Also, I'm measuring the above voltages from IG to OL/R... I'm not sure if this is the correct way to do this.

edit: ok, to OPAL I have 4.84,-4.85 and to OPAG I have -5.30,5.35 and to OPAR I have -5.55,5.54. Also, I tried to enable the bass boost, and the DC offset voltages are still disproporinate... now, with the knob in the off position, I get 1.9mV DC offset in the right channel and with it fully turned I get 7.4mV DC offset. In the left channel I get 0.5mV and 6.06mV. I also measured both channels of the pot, and they are a little off, the right channel of the bass boost pot is 1.4-45.8k and the left channel is 1.6-47.0k. Also, R4R = 10.0k and R4L = 9.98k.

I just ordered my Elpac, so unless there is something I can do with the 9v batteries, I'm going to retest after I get the regulated wall-wart power supply.

-Marc

 

[raduray]

I have a LaRocco PPA which is a PPA v1 board with the LaRocco Diamond Buffers. It's powered by an Elpac 24V Power Supply. Currently it's running OPA627 op-amps and I'm wondering if it will support AD843?

 

[quicksilver96]

Yes, an AD843 will work in place of an OPA627.

Edit for luvdunhill -

About the DC offset: small differences like this can be attributed to manufacturing tolerances. The difference you have described is very small and will not effect the operation of your amplifier. Some opamps such as the AD843 and OPA627 have pins dedicated to setting this offset exactly the same for applications where a few mV of DC offset will matter. The PPA (v1 or v2) does not connect anything to these trim/offset pins and there really is no need. If the offset difference is really high, there is something wrong with the build or the opamp itself.

About the input voltages: again - manufacturing tolerances. Tangent has an individual TLE, JFETS, R8 resistors, C4, and C5 for every opamp. Each of these parts has certain manufacturing tolerances and the V+/V- output to each individual opamp will be slightly different, though 1/2 volt seems slightly high, I don't think it will effect the operation of the amplifier. The "virtual ground" ouput of all the TLEs are tied together, and I think this is more important than the V+/V- levels being exactly the same at each opamp.

 

[tangent]

Quote:

Originally Posted by luvdunhill Do I solder on the front side of the board? There's no way I can really do this with all the parts already on the board..

I doubt there's really "no way" to do it, but no, that's not what I had in mind. With hookup wire, if you simply don't push the wire all the way in the hole so there's a gap between the insulation and the pad, you can heat the wire+pad up and flow solder in there from the same side as the wire.

Quote:

I have no idea what the G4 means

I haven't bothered to look, but I'll bet it's a RoHS code...Google it if you don't know about RoHS yet.

Quote:

What could cause a different DC bias in one channel but not the other? I have 2.2mV in the right channel and 0.5mV in the left channel.

First, the correct term for that is a DC offset, not a bias. EDIT: The distinction is one of intent: an offset is an error, whereas a bias is created purposefully.

As for what causes it, basically it's a difference in part tolerances. It's exceptional to have two circuits have the same DC offset, if you don't take special measures to ensure it. Mismatch is the common case. But as long as they're all within some reasonable range, it's fine, which is your situation.

Quote:

Originally Posted by raduray I'm wondering if it will support AD843?

We had problems with the 843 in both the v1 and in the v2. Do a search here...tips for increasing the amp's stability margin have been posted several times. A good first search would be for PPA and AD843.