luvdunhill Posted April 16, 2009 Author Report Posted April 16, 2009 Hm, I hate to offer this suggestion at this point, as it's not really a layout issue, but there may be benefit from splitting up the CCSes for the diamond buffers from one per rail on the original schematic, to two per rail, each one being dedicated for leg of the two diamond buffers. The slight improvement would be the fact that you don't have to worry about matching transistors, or noise from mismatched transistors. The only addition would be two more LEDs and an extra 30K resistor to the schematic.
DigiPete Posted April 16, 2009 Report Posted April 16, 2009 Marc, Once we settle on a schematic and you've done the prototyping, I'll use whatever comes out of that for the group buy passive parts BOM. It may be advantegeous to open up the lead spacing for the big resistors at this point, if possible. Pete
luvdunhill Posted April 16, 2009 Author Report Posted April 16, 2009 (edited) I think 512-MC78L15ACP and 512-MC79L15ACP are the best regulator choices. The 512-KA7915ATU and 512-KA7815AETU might be a second choice. They are TO-220 and handle more current, but they have 35V maximum input, where the above TO-92 parts have a maximum input of 40V. I actually couldn't find any fixed, TO-220 regulators with anything more than 35V max input, and the TO-92 regulators seemed to only go up to 40V. Anyways, I still recommend using the bigger opamp, but I suppose I can test these with a OPA134 just for the sake of testing. We also need to add a PDIP-8 socket to the BOM. Edited April 16, 2009 by luvdunhill
luvdunhill Posted April 16, 2009 Author Report Posted April 16, 2009 I just summarized all the above requests for change and PMed Jacob
digger945 Posted April 16, 2009 Report Posted April 16, 2009 It looked to me that the output resistor pads(R39 - R46) had two sets of holes, making it possible to stagger two resistors using 14mm lead spacing, one on top of the pcb using the first and third holes and one on the bottom using the second and fourth.
luvdunhill Posted April 16, 2009 Author Report Posted April 16, 2009 It looked to me that the output resistor pads(R39 - R46) had two sets of holes, making it possible to stagger two resistors using 14mm lead spacing, one on top of the pcb using the first and third holes and one on the bottom using the second and fourth. yes, or to piggy back the resistors on top of each other on top of the board. Now that I think about it, perhaps this is fine, but I think they could be moved slightly apart.
luvdunhill Posted April 20, 2009 Author Report Posted April 20, 2009 I have a question for the Spanish folk Do you ever place orders to Conrad.de? If so, would it be possible to add an item to an order? I cannot find a way to send to the US from their website.
Torpedo Posted April 20, 2009 Report Posted April 20, 2009 Not myself. Maybe Picaudio has, I'll tell him to come by
PICaudio Posted April 20, 2009 Report Posted April 20, 2009 I never seen this web & I don't need anything of this site at this moment. -- o -- Rodrigo
luvdunhill Posted April 20, 2009 Author Report Posted April 20, 2009 I never seen this web & I don't need anything of this site at this moment. -- o -- Rodrigo Thanks Rodrigo I'll see if I can find someone else to help.
Torpedo Posted April 20, 2009 Report Posted April 20, 2009 Mate, if you can't get it from anywhere else or they don't ship to USA, just say, I can order the items myself and send them to you
utilisateur Posted April 22, 2009 Report Posted April 22, 2009 I'll see if I can find someone else to help. Unless someone is shipping you something else, i had a one or two things i wanted to order i think, i'm in Germany for atleast1-2 more weeks so if you want just contact me =) They dont seem to have any free shipping actions running at the moment but theres still a small chance to hit their "every 85th order is free" thing EDIT: just looked and in fact they dont ship to the US and Canada
luvdunhill Posted April 22, 2009 Author Report Posted April 22, 2009 thanks guys, I'm talking to a US distributor and trying to pretend to be a company Also, looks like PcX sale that includes 20% off resistors is coming at a perfect time for us!
spritzer Posted April 22, 2009 Report Posted April 22, 2009 But, but... you are a company?!?! Marc-E-marc Audio and don't you forget it!!
luvdunhill Posted April 24, 2009 Author Report Posted April 24, 2009 Input transformers have arrived. If you would like yours *before* the semiconductor kit, please let me know. You'll have to pay shipping and packaging twice, but it won't be more than a few dollars more probably. Just let me know.
luvdunhill Posted April 24, 2009 Author Report Posted April 24, 2009 I was wondering if Pars or Jacob or perhaps someone else with Dynahi boards could answer something for me. What is the maximum width of a thermal pad that could be used and there not be any overlap between the pads? Not that this is bad, but looks kinda funny and is wasteful. I'm talking with a company that makes isolation pads that perform orders of magnitude better than the products available at Mouser / Digikey. To give you an idea, thermal conductivity of 8.0 W/mK and thermal resistance of 0.07 K/W. This is around 8 times better than for example Thermalsil III... and so far I'm seeing around a 8 degree temperature drop which is consistent with this. This is rather amazing result, IMHO (think about how much larger heat sink you would have to use to get this sort of drop!)... Since we're going to have to order pads anyways for the part kits, I was considering using these ... I'd have to get them custom cut, and there is a lead time, but I think it would be worth it for those interested.
digger945 Posted April 24, 2009 Report Posted April 24, 2009 I was wondering if Pars or Jacob or perhaps someone else with Dynahi boards could answer something for me. What is the maximum width of a thermal pad that could be used and there not be any overlap between the pads? Not that this is bad, but looks kinda funny and is wasteful. FET pcb image on bottom, DHi pcb on top, Renesas 78/215's. Lemme know if someone decides to back out on the inputs.
luvdunhill Posted April 24, 2009 Author Report Posted April 24, 2009 (edited) Those have the older package markings, they're probably Hitachis (or an eariler non-RoHS version I suppose). Anyways, so, does that mean there is 0.02" between the two devices? wowza, I would have guessed more. Thanks again! Edited April 24, 2009 by luvdunhill
luvdunhill Posted April 24, 2009 Author Report Posted April 24, 2009 (edited) what about a 12mm wide pad? datasheet for MOSFETs with case dimensions: http://documentation.renesas.com/eng/products/transistor/rej03g0122_2sj76ds.pdf Edited April 24, 2009 by luvdunhill
digger945 Posted April 24, 2009 Report Posted April 24, 2009 My bad. Yea .026 between the metal tabs, about .068 between the bodies. The body itself is .3995 wide and the tabs are .4130. They don't have Hitachi marking like others I've seen online.
luvdunhill Posted April 24, 2009 Author Report Posted April 24, 2009 (edited) My bad. Yea .026 between the metal tabs, about .068 between the bodies. The body itself is .3995 wide and the tabs are .4130. yeah, the datasheet says up to 11.5mm max tab diameter, see how they jut out a bit further in? The pad widths are available in 1mm increments, so 12mm, 13mm, or 14mm would be the reasonable choices. Better than hand trimming the pads, right?! I'm betting they are simply non-RoHS devices. The ones I have were dropped shipped from Renesas and have different case markings. I'll post a pic sometime soon. Edited April 24, 2009 by luvdunhill
digger945 Posted April 24, 2009 Report Posted April 24, 2009 I measured and compared to the datasheet you linked to and everything is within specs. The measurements without a tolerance are all exactly spot on. 12mm looks too wide to prevent overlapping but will fit. 11mm would fit better given the current spacing. My dynahi pcb lays exactly on top of the printout I have and is exactly the same width with .3" more depth.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now