[tommyatkins]

EDIT: Dang this looks long, but I promise it won't take that long to read. Most of the real estate is parts.

So here is my list of parts for my PPA. I have a few questions which aren't particularly necessary to have answered, but maybe it'll help others out. I know I found cyntax's list and following discussion helpful. Most parts are right from Tangent's guide. There are a few variations, and I had a few questions, so I decided to post in the hopes that it will be helpful to others, as well as to me.

I often note the prices of things, or particular options. When I do so, I am usually talking about the total cost if I went with that option (ie. to change both C2 and C5 would take a total of 9 caps, so I list the total price of 9 caps). I give these types of details hoping that it might help someone who is in a similar situation to me. I haven't seen much in the way of specific prices linked to particular part selection. I've done a good deal of reading, thinking, and weighing/comparing prices, so maybe someone will find this useful.

(All resistors are Vishay/Dale 1/4W 1% Metal Film Resistors unless otherwise noted)
R1 (3) - 4.32K
R2 (3) - 1M
R3 (3) - 1K
R4 (3) - 10K
R5 (2) - 3.32k
R6 (2) - 1M
R7 (2 each) - 10K, 15K, 22K, 47K, 75K, 100K (for experimentation)
R8 (3 each) - from 10 - 1k of KOA Speer 1/4Watt Axial Leaded Carbon Film Resistors 5% BULK (yes a lot, 46 values in fact. Total = $6.90. I'll find uses for the ones I don't use).
R9 (3) - 1K
R11 (3) - 1K
R12 (3) - 1K
R13 (3) - 1K
R14 (3) - 1K


Maybe R8G (1) - 1M (Didn't I recall Morsel saying the R8 on the ground could be left open? I was just going to buy an extra one, in case I found that post again. It's only $0.06)

C1 (2) - Nichicon UPW Radial Aluminum Electrolytic Capacitors 35V 1000uF 12.5x25 (Mouser: 647-UPW1V102MHH)
C2 (3) - Vishay/Roederstein Metallized Polyester Film Capacitors 1.5uF 50V 5% (Mouser: 75-MKT1826515054)
C3 (3) - Vishay/Roederstein Metallized Polyester Film Capacitors 0.1uF 63V 5% (Mouser: 75-MKT1826410064)
C4 (6) - Nichicon UPW Radial Aluminum Electrolytic 35V 100uF 8X11.5 (Mouser: 647-UPW1V101MPH)
C5 (6) - Vishay/Roederstein Metallized Polyester Film Capacitors 1.5uF 50V 5% (Mouser: 75-MKT1826515054)
C6 (1) - Kemet Radial Ceramic 200V C0G 10pF 5% (Mouser: 80-C315C100J2G)
C7 (2) - .12, .091, .075, .068, .051, .039 uF BC Components Metallized Polypropylene CAP FILM MKP 63VDC 2%

Q1 (3) - Fairchild 2N5484 Transistor (Mouser: 512-2N5484)
Q2 (3) - Fairchild 2N5486 Transistor (Mouser: 512-2N5486)
Q3 (12) - Fairchild 2N5486 Transistor (Mouser: 512-2N5486)

D1 (1) - Rectron Rectifiers, Silicon DO-41 1A, 50 Prv (Mouser: 583-1N4001)
FET (1) - Fairchild 2N5484 Transistor (Mouser: 512-2N5484)
ZNR (1) - ? (see below)
LED (1) - Blue (Mouser: 606-CMD264UBC) or Amber (Mouser: 512-HLMP1719) Probably Amber, I'm traditional (but not red traditional

)

S1 (1) - E-Switch Rocker Switches DPDT ON-ON PNL 10A (Mouser: 612-RR3130B) - Yeah Rocker Switch (Seriously, I'm stoked about this part)
S2 (1) - E-Switch DPDT ON-ON SLDR MNT (Mouser: 612-100-F2311)

Out (1) - Neutrik Phone Jack 3C 1/4" CHASSIS-JACK (Mouser: 568-NJ3FP6C)
IN (2) - DGS RCA Phono Jacks, Nickel: Red and Black (Mouser: 161-2002, 161-2003)

Sockets (23) - Mill Max (Mouser: 575-113308)

Molex KK 3C HEADER (Mouser: 538-22-23-2031)
Molex KK 3 CKT HOUSING (Mouser: 538-22-01-3037)
Molex KK CRIMP TERM 22-30 TIN (Mouser: 538-08-50-0114)

Power Jack: Deltron DIN Connectors JACK 5 PIN BLACK (Mouser: 164-2521) (EDIT: I had forgotten this in the original post, so I include it for reference, but I am not going to go this route)

Power Supply: Elpac WM071-1950-D5 7.5 Watts 15V/-15V (Mouser: 680-WM071-1950-D5)

Enclosure: Hammonds 1455n1601BK, Black with Metal Panels (Mouser: 546-1455N1601BK)

I'll order all I can from Tangent. Hard to beat the prices, and I know the very little profit he makes on this helps off set his cost of development (not to mention the all the time he and the other put into the project).
OPAMP (3) - AD8610 mounted on Adaptors
VOL (1) - Alps Blue, 50k
TLE (3) - Texas Instruments TLE2426CLP ("rail splitter")
PPA PCB (1) - It kinda makes the process a bit easier.

Finally Buffers from Newark (Arrow seems to be out, I will call though):
BUF (12) - Intersil HA3-5002-5 (Newark: 06F5241)

Totals (not including shipping):
Mouser = A squigen under $130
Tangent = $68.53
Newark = $46.20



R7 - Is this a good mix, for experimenting with bass boost values?

C1 What is the maximum clearance in the Hammonds case if I want to use a battery board? From the pictures, it seemed like 25mm should work. I didn't notice much on this (I couldn't find Doobooloo's height of his too-tall C1 caps).

C2/C5 - I can't really decide here for every .5uF (over 1uF it's about an additional $11. I'm thinking I'll go middle ground with the 1.5uF caps. Any input?

C3 - So I only add this if the amp isn't stable right? It's better to have it out? I was just going to order both .01uF and .1 uF, to have both options, if I needed to. It's only $0.63 for each set.

C7 - This one part's gonna cost around 15 bucks with shipping, and a $5 handling charge. I may shift some of the parts from Mouser which cost almost the same to boost my order above $25, and avoid the handling charge. (assuming the net difference is less than $5...).

ZNR - Not sure what the different voltages mean, I'll research this. Also, I'm not sure on how many batteries I'll have (I'm going to make the batt board when I have some more money some time). I had planned on purchasing all the Zener Diodes Tangent suggests (20 of 'em, but only like $0.80 total)

I may go with some other input jacks (Gold methinks). I might also give it two inputs (All I'd need is another DTDP switch - like S2 - and two more RCA Jacks, right?

I am going with the 30V Elpac as I think I will need this when I add the battery board (based on the measurements I've made from the Tangent's guide). This gives me a bit more flexablity later on.

I think I have selected the correct power jack. From the data sheet, it seems the 15v/-15v Elpac uses a 5-Din connector.

Not sure if I'm going to use the Hammonds Case, I might build a wood one, or just paint the Case all fancy-like (I can't remember who, but some head-fier posted a really fun looking one)

Ok I wasn't sure what the Molex Crimp Terminals are for, but I figured it out as I was typing (I think) so I'll include it for others who might have questions. Only $0.21, so I was just going to figure it out when I got it. As I understand it the Molex KK is so the wire for the DC jack isn't directly attached to the board. The Molex KK system is where you buy a header (which goes on the board? or maybe its the housing) and a housing (which goes on the wires?) but you purchase the number of terminals (ie wires to be put in the housing) separately, then assemble the housing and terminals. This is so you can actually attach the wires to the terminals easily (otherwise space is cramped).

I will also be adding a crossfeed board (the one from Tangent) this will attenuate the signal, but I don't really need to adjust any of my parts right?

Please make any other comments on part selection, if you wish.

FYI, I play on listing with my AKG K501s to my computer, Sony D-25s, and some old JVC TT with a Nad PP-1 Phono Preamp.

Thanks to the whole PPA team, great work (I heard a prototype at the June Bay Area meet), and thanks for the help.

(I've previewed this a bunch of times, ironing out some errors, and formatting. It's 2 am now- Happy Halloween!- and I have class tomorrow, so goodnight, and I hope it's readable).

 

[tangent]

Quote:

Didn't I recall Morsel saying the R8 on the ground could be left open?

You must be thinking of R3.

R8 is part of the current source, so if you left R8 open you'd leave out the JFETs and R9 as well. If you did anything nonstandard with R8, you'd jumper it, but that would require hand-matching JFETs, a more expensive proposition than buying a resistor assortment.

Quote:

Yeah Rocker Switch (Seriously, I'm stoked about this part)

The only hitch on that is being able to make a 3/4" hole. I use a panel punch. I wouldn't want to try using a drill, unless I had a drill press and clamped the panel to the table.

Quote:

Elpac WM071

I wouldn't go that way.

For one thing, this is a true dual-voltage supply. Either you'll end up ignoring its ground pin or you'll have to hack up the virtual ground setup on the board to make the power supply set the signal ground level. The PPA was designed to be run with a single-voltage supply.

For another, it's 30V rail-to-rail, which will destroy AD8610s. There's the related +/-12V WM072, but it would be cheaper to go with the WM080 which puts out the same voltage rail-to-rail.

Also, I don't see a DC input jack in your parts list. For the 07x family, you'd need a DIN-5 female jack. (This is another 3/4" hole (plus or minus), incidentally.) If you go with the WM080 I recommend, the DC power jack in the parts list is ideal.

Quote:

R7 - Is this a good mix, for experimenting with bass boost values?

You've got a mix of ~1.5x and ~2x jumps in there. I'd either prune the list to a simple 2x series (10K, 22K, 47K and 100K), or also add in 33K to complete the 1.5x series.

Quote:

C1 What is the maximum clearance in the Hammonds case if I want to use a battery board?

25mm for components near the edge like the pot and the C1s. It's more like 31 or 32mm out in the center. This assumes that you're using AAAs. I'm not sure anyone's tried AA arrangements yet, but that would eat an additional 3-4mm.

Quote:

C3 - So I only add this if the amp isn't stable right? It's better to have it out?

It's not going to hurt performance to add it. It's just a cost issue governing whether to leave it out or not. You'll notice you haven't yet heard anyone saying they had to add C3 due to instability.

Quote:

C7 - This one part's gonna cost around 15 bucks with shipping, and a $5 handling charge. I may shift some of the parts from Mouser which cost almost the same to boost my order above $25, and avoid the handling charge. (assuming the net difference is less than $5...).

You might consider getting OPA627s or 637s from DigiKey for the op-amps instead of the 8610s. This will get you over the low order limit. You'll spend more, but I'm having misgivings about 8610s right now personally. I haven't pinned it down to the chip itself or the sources yet (too little time) but I can say the 627/637s are...shall we say "less adventurous?"

I'm running a 627/637 PPA with a 12xAAA battery board being charged from a WM080. Works just fine. I might wish for a bit more voltage to the op-amps (hence more batteries, hence a higher charging voltage), but I personally would rather use a custom power supply rather than go with a +/-15V supply wired rail-to-rail.

If you can't wait for me to get around to finishing the PPA PS project (which will exist to fill this very need), I could help you replicate one of my current prototypes. They're LM317 based, and they perform better than the Elpac WM080. The only reason I'm not steaming ahead with this design is that I'd like to try a better design and do some more testing first. The parts are probably in the $60 range.

Quote:

ZNR - Not sure what the different voltages mean, I'll research this.

I just rewrote the similar section in the PIMETA docs. Would you please read through that and see if it helps clear things up? If it does, I'll consider reworking the text in the PPA docs. The two say essentially the same things, but in different ways.

Quote:

I may go with some other input jacks (Gold methinks).

Radio Shack has gold RCAs in the same (cheap) style as the DGS ones from Mouser for $3 per pair. If you want something better quality than that without getting excessive on the price, I'd recommend shopping at Parts Express. They have a few different higher-quality RCAs, and they include insulating washers.

Quote:

I might also give it two inputs (All I'd need is another DTDP switch - like S2 - and two more RCA Jacks, right?

DPDT, yes. If you did it that way, you'd be grounding the two sources together all the time, which will usually work fine. (Your preamp probably works this way, too.) If you do have grounding problems you could go to a 3PDT switch and switch the ground connections as well. That switch will just be a bit wider behind the panel, so if you leave enough room for it, there won't be much risk in going with the DPDT from the start.

You'll have many long wires from the front of the case to the back, so observe good wire harness technique to avoid degrading performance. Consider testing the amp with just one set of jacks wired to the rear 'I' pads, then reworking it to a dual input later. If it performs worse after moving to dual inputs, you'll know why.

Quote:

I wasn't sure what the Molex Crimp Terminals are for

You crimp them on the ends of the wires. The female KK connector has no contacts of its own. You crimp a terminal onto the end of each wire, then push them into the connector and the terminals click into place, ready to mate with the male KK connector.

Quote:

I will also be adding a crossfeed board (the one from Tangent) this will attenuate the signal, but I don't really need to adjust any of my parts right?

You might get some extra 15K and 22Ks for R4, in case you need to raise the amp's gain to compensate. 10x gain is often plenty, even with crossfeed.

 

[Voodoochile]

Quote:

Originally posted by tangent The only hitch on that is being able to make a 3/4" hole. I use a panel punch. I wouldn't want to try using a drill, unless I had a drill press and clamped the panel to the table.

I have a very nice drill press, and a full set of bits to 1/2", plus a few larger sizes with undercut shanks.

That said, even with that equipment drilling a 3/4" hole is a significant undertaking. you can do some serious damage, nut just to the panel, but your wrists, if not done very carefully. In other words, it goes beyond having the hardware.

The punch is the way to go. If no punch, then a stepped bit is the next easiest. Using regular bits, there are a variety of ways to make an accurate hole, but they are somewhat invloved, and potentially hazardous.

I'm not trying to scare anyone off, but if someone were to think "I've drilled beautiful 3/8" holes in metal before, I'll just get a bigger bit.", they will be in for a surprise, I believe.

 

[phaedrus]

My best casing investment to date has been the pair of stepped bits I purchased -- the second best may be the drill press currently sitting unassebled in its box on my garage floor. My large stepped bit only goes to 7/8", which is shy of what a Neutrik locking 1/4" headphone jack wants. A little patient work with a deburring tool handles the fit for these monster components.

Also, after trying my first couple of holes in the hand held style of the poor DIY college student, I spent $15 at that orange home improvement store for a drill press vice. I'm still drilling with a hand held drill, but not having to manage the panel at the same time. Oh, and using a punch on the center point of your panel holes goes a long way to reducing drift.

 

[A3rd.Zero]

one of the big problems with large bits and large holes is getting a perfectly round hole, the bit tends to mash up the hole real good. If you dont have a stepped bit a fullsized bit will work but you must drill a piolt hole with a smaller bit, say 1/8". Also if you dont have a drill press you can take the plate of aluminum and screw it into a sandwitch of plywood. if you can bolt through the ply and through the mounting holes of the plate that is the best. Then just clamp the plywood and drill slowly! youll want to use a cutting oil of some sort.

to make sure that your holes line up, drill the plate for the 1.8" hole and plywood for the bolt holes, then mount the plate on top of the plywood and drill the 1/8" hole in the plywood. then sandwitch and go nuts (well go nuts slowly). there is a misconseption that drilling faster will make a cleaner cut.

NOTE: if you are a college student, likely your college has shops for Art students, Theater students, engineers, and whatnot. Friends who can get you into a fully functional, or even partly functional shop are worth their weight in gold.

 

[tommyatkins]

Quote:

You must be thinking of R3.

Ah, right. I was kinda confused after reading your docs and such. I know not everyone on the PPA team agrees on everything 100%, but that seemed way off.

Quote:

The only hitch on that is being able to make a 3/4" hole.

Yeah, not positive how that will work out. I might be able to get into a shop, or I might just case it in wood (with one of the shielding options mentioned in one of the wood case threads). Or I might just use a toggle with enough space to enlarge to a rocker when I find someone with the right tools. I read some of the older threads regarding making holes in cases when they were first posted, so I am aware of those issues. I solve the problem somehow.


Quote:

Re: Elpac WM071 I wouldn't go that way.

I had forgotten I will be able to switch to the custom powersupply later (probably when I get around to the battery board anyways). And I hadn't noticed that it's not really compatible

. Thanks for the offer on the prototype info, but I'll probably not have enough money/time for the battery board for some time anyways. I am looking forward to it though.

Quote:

Also, I don't see a DC input jack in your parts list.

Oops, I knew I forgot something. I had intended to use:
Deltron DIN Connectors JACK 5 PIN BLACK (Mouser: 164-2521)

Quote:

You've got a mix of ~1.5x and ~2x jumps in there. I'd either prune the list to a simple 2x series (10K, 22K, 47K and 100K), or also add in 33K to complete the 1.5x series

I kinda just played around with a spread of values, Since there is larger jumps in boost between the lower values, I weighted that end of the values. But it would be nicer to have a complete exponential series. I'll probably get a 33k

Quote:

You might consider getting OPA627s or 637s from DigiKey for the op-amps instead of the 8610s. This will get you over the low order limit.

Ok, I might do that. I wanted to get both the AD8610 and the OPA637/627s at somepoint to roll 'em. That's $20 more, but I guess only $15, since I avoid the handling fee. Actually these might be a better match for my AKG K501s from what I gather from impressions here.

Quote:

Re: ZNR selection I just rewrote the similar section in the PIMETA docs

Ok, it makes a bit more sense now. Ok, I reread part of my Radioshack "Getting started in Electronics" and I think I understand now. The voltage ratings listed in the parts list is the reverse breakdown voltage, right? (also known as VsubZ?) So I choose a value combined with the the LED voltage (Maximum forward voltage on the datasheet, yes?) is my minimum working voltage. I'll get Zeners one step up and one step down, as you suggest also.

The most confusing thing for me is what voltage ratings are what. I understand many of the general things happening with the circuit, and specific parts, but I sometimes get confused with terminology as it isn't always completely consistant. The PIMETA ZNR description helped because it told me to look at the datasheets again. I already understood what it was saying though (from the PPA description, and reading the minimum working voltage section). I think the PIMETA would help more if one was unfamiliar with batteries and such. The section on minimum working voltage helps give another angle on the Zener. I would put a link in both the PIMETA and the PPA ZNR section linking to each other, with some comment like "Still confused? Maybe this look at ZNR selection will clarify things." Like you said, they say the same thing, but different. Again, what hung me up was being able to read the datasheets, not your tutorial (except I didn't know where you got your numbers, but if I understood the datasheets better then I would have known). I think I get it now though. (do i?) I just keep reading and rereading datasheets and information, it's starting to soak some more. Thanks for the help.

Quote:

If you do have grounding problems you could go to a 3PDT switch and switch the ground connections as well

Ah, yes. I always get confused on switches without having them infront of me. Right, the third ground would be better.

Quote:

Re: Molex KK You crimp them on the ends of the wires.

Got it, thanks for clarifying.

Quote:

You might get some extra 15K and 22Ks for R4, in case you need to raise the amp's gain to compensate.

Will do. Thanks.


Thanks again for the help, Tangent and others.

 

[jamont]

Quote:

Originally posted by tangent If you can't wait for me to get around to finishing the PPA PS project (which will exist to fill this very need), I could help you replicate one of my current prototypes. They're LM317 based, and they perform better than the Elpac WM080. The only reason I'm not steaming ahead with this design is that I'd like to try a better design and do some more testing first. The parts are probably in the $60 range.

I'm thinking about building a single-supply PS instead of buying another Elpac, and I'd be interested in seeing the details of your LM317 based prototype.

 

[tangent]

Anyone else want to see details on this supply? Also, will you want board art to etch your own boards, or do you just want part lists and a schematic so you can make it on protoboard? I need to know so I can decide how much it will derail me to put this info up in a comprehensible fashion.

 

[jamont]

I can get by on a schematio and parts list, since I'll probably build on perfboard. But it would be helpful to see your layout as well if it's not too much trouble. Thanks!

 

[Gariver]

Tangent:

Can you elaborate a little bit on your panel punch? It sounds like a great tool to have. Is it very expensive? TIA!!!

 

[phaedrus]

tangent, I'm interested in the PS and would lean towards etching a board (haven't done that yet) if this is a good learning project for that. I may build it up on protoboard first, for the exercise.