SSL9K parts questions - working BOM

just a few questions - I am throwing together a BOM:

What is the value of the 2 caps labeled "CX"?

Also, there is no value for caps 47 and 74 on the main board given in the schematic

The dual transistor - does anyone have a part number for this or mouser part number?

regarding using diodes to step the +/- 18V to +/-15V, should I just use a standard rectifier diode for this? I have heard of people doing this before, but I am not sure what kind of diode to use.

lastly, what are people using for caps 55,68,69,84?



I was unable to load Neeno's BOM for some reason, and there is not a ton of info about this build. I will continue to search but if someone knows the answers I would appreciate it.

thanks!

[quote author="Milkmansound"]What is the value of the 2 caps labeled "CX"?[/quote]

pretty sure those are 100nF (0.1uF)

[quote author="Milkmansound"]Also, there is no value for caps 47 and 74 on the main board given in the schematic[/quote]

should be 3.3pf, COG or NPO

[quote author="Milkmansound"]The dual transistor - does anyone have a part number for this or mouser part number?[/quote]

MAT-02, available from Analog Devices. Digikey part number.... MAT02FH-ND. MAT02EH-ND is also ok, but more expensive (and IIRC, not all that much better). They were out of stock for a long long time on these, but they appear to be carrying some again.

[quote author="Milkmansound"]lastly, what are people using for caps 55,68,69,84?[/quote]

I think wima's should fit nicely there, that's my plan anyway, although somewhere there are some threads where andy millar graciously gives some thoughts on parts selection for these boards, and he said that for the most part the quality of many of these caps is not paramount, although the 2.2uF may have been the exception. I can't recall exactly, but I do believe that C68 and C84 are phantom blocking caps, and possibly the most crucial place for quality caps.

Either way, WIMA's should fit well in circuit and in layout.

As for the diodes, I plan to use this method as well, and I think that this is discussed in the most recent 2-5 pages of the main ssl9k thread.

Also, someone please correct me if I'm wrong about any of this. Also, I'm not sure how quality the 2 inductors should be. They may also be in the circuit to keep it's bandwith very high (one the main design goals for ssl)

billy

edited for /quote usage idiocy

thanks, by the way, for working on this.... definitely a very useful thing to have.

oh, by the way, I believe that pin spacing for the 2.2uF caps is different for each... the 63v phantom caps, C69 and C55, look to me like 15mm spacing, although i don't have a board and calipers handy to double check. C84 and C68, also 2.2uF, look to me like 10mm spacing. The others are I believe 5mm.

i found out my local surplus place has some wimas on the shelves, so I was just going to bring a board with me and tool around trying to fit the right wimas ;]

thanks again

billy

use a diode with a high drop, like a standard silicon. drops .5 to 1.5v depending on the maker etc. two 1.5v drops will give you 15v..

the only wima's in the mouser catalog that are 2.2uf look fairly huge:

W11 H21 L26.5mm wowsers - I think they will still fit into a 1U case, and there seems to be room on the baord - but man, thats pretty huge no?

I am not having any luck finding TL051's either... Digikey has some SOIC TL051's, and I can adapt using a soic to dip adapter - but first I want to see if I can find these. Will anything else work in there instead?

I will definitely post this BOM when I am done if people want it.

oh, and of course its not easy to find these dual rev logs pots either - using the rotary switch is an option for me, but 6dB steps seem a little extreme.

this whole build is going to be extreme!

yeah wimas can be tough to find, at least the right wima for the job. it may be a better idea to use a more standard film cap for the bom, and those that can find the right wima's are the same people that will want to use them. i've used them in a lot of places, but mostly lucked out as far as getting the right physical size.

tl052's are the way to go for those ic's. plenty of them around. tons. and they'll work fine, and are also in the neeno bom. theres a group buy in progress for the pots, from omeg. they arent so easy to find, especially in the states. farnell or newark or one of the british companies may be the way to go if not a group buy from omeg.

i was going to use the rotary switch, but i too would like to have a mic amp with a pot for once.. i have switches on my green pres, and my 312's and 1272 in progress are both getting switches... thats why i hopped on the group buy.

thanks for kicking in about the diodes svart as i've been wondering what to use there as well, and somehow, after reading the original idea months ago on the ssl9k main thread, what you just said just helped me realize more of whats up with diodes

according to the texas instruements data sheet, the pinout for the TL051 and TL052 are completely different, whereas the 51 is a single and the 52 is a dual without the offset on pin 1.

are we sure this is a direct replacement?

I decided to try these caps cause they are much smaller - mouser P/N:

5989-100V2.2-F

should be ok I think - and they will at least fit!

http:www.milkmansound.com/img/SSL9Kparts.xls

can somebody look this over for me?

I am probably going to do a little juggling around to eliminate one of the suppliers if I can. This was not easy to put together - and its still missing a solution for the dual log pot, but its a start.

thanks

Any particular reason you're listing those NE5534s? The 595-NE5534AP can be had for nearly half as much.

Matt

[quote author="Milkmansound"]I am not having any luck finding TL051's either...[/quote]
You don't need '51s - the schematic is wrong. "IC31a" is actually a '52 and "U32b" should have been listed as "IC31b" which is the other half of the TL052. In other words, just take the existing schematic and change TL051 to TL052, and "U32b" to "IC31b". I don't have a copy of Neno's BOM either, but it's probably from European sources so you'd still have to track the parts down for US.

By the way, the +/-15V supply is just for the TL052. I guess it's just playing it safe since the '52 is rated at +/-18V max. You could kludge one or two diodes in series between the +/-18V and +/-15V pads to drop a little voltage (Observe polarities! Also, there's a thread on that, but I don't know if it made it to the meta...), or you might could just connect the +/-18V pads to the +/-15V pads and use an OPA2604 which is rated for +/-25V. The reason I say "might" is because no one has tried this to my knowledge. I don't see any reason why it wouldn't work, though, albeit a bit more expensive.

noted and updated

thanks

[quote author="Flatpicker"]could just connect the +/-18V pads to the +/-15V pads and use an OPA2604 which is rated for +/-25V. The reason I say "might" is because no one has tried this to my knowledge. I don't see any reason why it wouldn't work, though, albeit a bit more expensive.[/quote]

very worth a try..... especially as i have some free samples of those still lying about.

i have some signetics 5534's i got from apex jr. for this one. great price as well, can't remember how much it was exactly.

but the bom looks good... only things i'd wonder about are the small bear stuff, as i believe mouser also carries the b series poly's. not only that, but i also think that they carry enough wima's to cover all but the 15mm 2.2uF's and the 10mm 2.2uF 63V. The latter they have, but only at high quantity. Lower values, like 100pf and 330pf and so forth are probably in polyethylene rather than polyester film, or you can just search for FKP, as that series includes the lower capacitance wimas.

nothing against the panasonic b series, they're fine as well. most of these arent super critical either. wimas can also be had pretty cheap around ebay however.... a bit outside the purpose of good BOM though.

thanks for the work on this stuff.... do check out some of the mouser stock of wimas, and the product selection guide on wima.com can be really helpful for figuring out the right series to look for. 470n wima's, for example are almost surely a bit cheaper than the silver micas.

anyway, good stuff. if i have some time tomorrow maybe i can help fill in an alternate wima section or something

billy

The new wima cases are a bit of a pig as the values for bottom left of the board are too thin/thick to fit side by side
I had to bend the legs - like a dog leg to get them to fit

As Simon has already said the Wimas (470nF 63V) do fit but some bending of the leads is required. Not sure if the photo below shows this or not. The 100nF Wimas all fit with no problem except for C63.

The 2.2uF 63V caps are Wima MKS4 and again need bending to fit properly. Also on Gustav's PCB the holes for the MKS4 need to be enlarged slightly for the leads to fit.





**edited to correct 'C83' to 'C63'

maybe those panasonic b series are the way to go then... long leads in case of problems like that. I seem to remember the cap's not being too much of an issue sonically in this circuit, as andy m. was saying....

[quote author="Andy Millar"]At first glance I don't think any of the capacitors are very critical for type in terms of audio performance (except the 2u2 input caps), most of them are there to achieve a dead flat high frequency response - we're talking 200kHz not 20kHz so any effects a dodgy dielectric produces should only be of interest to highly trained bats. As suggested the smaller ones should be ceramic. The 470n can be anything you like - they're to produce the DC 'servo' voltages so they don't pass audio. The 330pf and 470pf I probably would use some sort (any sort) of plastic film type, but to be honest I would be very surprised if anyone noticed if they were ceramic. But I'll have a bit more of a think when I'm awake.[/quote]

[quote author="Andy Millar"]The 3p3s, 10ps and 100ps don't need to be NPO/COG, any old ceramic will do, for highest performance the others up to 470p could be. (Because NPO/COG capacitors are very stable with temperature this stops any high frequency phase differences between the two halves of the balanced channel. But it is probably me being a bit perfectionist!)

In answer to the question way above, the reason for using ceramics is (to put it simply) they keep their capacitance up to very high frequencies - well into the MHz - and the challenge with 5534s is to stop them turning into radio transmitters which is what the really small value capacitors are for. Otherwise ceramics are dreadful for audio due to their distortion, but this is only a problem if the capacitance is starting to have an effect seriously down into the audio band, which none of these are.[/quote]

so maybe cheap ceramic for the 3p3, 10p, and 100p, and cog/npo on the 330p and 470p

cheap poly on the 470n, as its just in the servo circuit. Maybe someone knows how crucial the 100n is in the circuit? (doesnt look very to me, just looks like its passing some freq's to ground, but i don't know)

i think the input caps are the challenge, cuz finding those spacings in the states is kind of ridiculous.

anyway, figured i'd look up that info and post it.... pretty damn cool to have advice from the source

billy

I would love to eliminate the small bear things and just do all mouser, but I was having a hard time finding things that were small.

If someone wants to throw in some wima part numbers that will fit, I will add them to the BOM that would be great!

I do not think mouser carries the panasonic caps - but digikey might. I will investigate.

Also realized that I need to add this 8 Dip sockets!

added B series panasonic caps from digikey - can someone please just double check those for me?

this is coming together nicely - thanks to everyone who is contributing. The US BOM for the 9K is all but done!

added in rotary switch and resistors, and a few other odds and ends

just need someone to glance over my work then I think we can call this one done!

I'll check for some cap alternatives and post what I find. I've found more discrepancies between the board and schematic. I'll probably just scan a marked-up copy and post that as well.