Newbie builds GroupDIY 51X rack

[[silent:arts]]

the original API L200 PSU uses 20AWG btw,
but not marine grade

 

[jsteiger]

...Patience is wearing thin on the official 51X Alliance PSU metal kit shipping to the US.  I haven't heard anything new about progress on that front even though I've been keeping a close eye on the thread for any new developments...

'Chunger, last I heard from Cemal, he was to pick up all the PSU cases at the end of this past week. If that did happen, I would gather that I will have those cases to disperse before you could get a lesser case from Parmetal. I understand if you want to move forward. I just saw this and figured I would let you know.

Cheers, Jeff

 

[chunger]

Thanks for the update Jeff. . . do you know if at least a partial kit is going to be offered (the items that physically mount on the case).  I have been searching for the power entry module (switched 120/240), the mains switch (rotary), and fuse holders (14mm hole), and have not been able to come up with a US based source for parts that match install holes.

If not, I guess I should see if RS ships to the US and how much that's going to run.

 

[chunger]

Is it physical size or rating that you are concerned about on the cable...that 24awg is rated at 300V and will carry the load perfectly fine in short runs.

You've gotta take this with a grain of salt because admittedly, I don't know what I'm talking about, but most wire I see (any gauge) is either 300V or 600V rated, but that doesn't tell you how much current it's rated to carry.

Double grain of salt because this is now internet knowledge discovered by someone who doesn't know what they're doing. . .

AWG  --  ohms per 1000 ft. --  max amps for chassis wiring  -- max amps for transmission
20 AWG -- 10.15 --  11A  -- 1.5A
24 AWG -- 25.67 -- 3.5A -- .577A

http://www.powerstream.com/Wire_Size.htm

My understanding is that we need 1.6A per rail for 51X.  For transmission, it seems even the 20 gauge is not quite up to standard but close, and the 24 is underrated.  If you consider the run short enough that for all intents and purposes, it's "in-chassis", then I guess the 24 works fine.  For something so foundational though I like to be completely above board.   If heat is an issue on the power supply, I can see myself wanting to install it in a separate rack perhaps some distance away from the 11 slot chassis.

 

[chunger]

Well.. . figured I should at least attempt to finish a project I start. . . so I ended up going to a local Fry's Electronics and picking up a few small do-dads to try and get my rack buttoned up while waiting for PSU parts.  What follows is probably not the best way to do things, but the store was out of straight headers, so I purchased some angled ones.  I probably would have just bought both if they were available.  I remembered reading that some folks has clearance problems with the straight ones on 1 particular 500 series module. . .

These were pretty easy to cut to proper length.

Then, I jammed the little plastic tab down with my thumbs

. . .and figured out I probably have a little bit of a clearance problem on the stereo link with this arrangement.

So, because I was in a bit of a hurry, I decided to "hack" the parts to work even though buying a proper straight header would have been the proper way to go.  I thought about it for a little while and decided it would be difficult applying the little shunts sideways with my hand deep inside of a rack, so straight up (as designed) seemed the be better for me.

. . . and, I have plenty of clearance on the back side with these for soldering.

These would have been much easier to solder in prior to the XLR connectors going in.  If building again, I would have definitely done it in that sequence, but a little patience got the job done. . . not the cleanest solders in the world but the meter tells me the joints are looking ok.

. . . and where I stopped last night.

I read conflicting reports on forum about which way to ground.  Silent-arts seems to advocate pin 1 to chassis while jsteiger seems to advocate pin 1 to ground.  Well, I guess that's why I put the headers in.  For now, I have pin 1 to ground, but I'll probably put pin 1 to chassis and then check to see if phantom power is working properly as my feeling I get from reading suggests this is a more "proper" way to do it.  If I get noise or phantom power problems, I can change it over to the other configuration without much problem.

 

[jsteiger]

...I read conflicting reports on forum about which way to ground.  Silent-arts seems to advocate pin 1 to chassis while jsteiger seems to advocate pin 1 to ground.  Well, I guess that's why I put the headers in.  For now, I have pin 1 to ground, but I'll probably put pin 1 to chassis and then check to see if phantom power is working properly as my feeling I get from reading suggests this is a more "proper" way to do it.  If I get noise or phantom power problems, I can change it over to the other configuration without much problem.

Volker is right on this. In this type of configuration pin #1 should go to chassis as per the Rane documents. The shunts and headers really make it easy to swap if need be but you should start out with pin #1 to chassis.

Cheers, Jeff

 

[chunger]

Just for my own info on the next step (just using resistor color sequence on Neutrik pins for power):

1 = Chassis - brown
2 = Gnd - red
3 = +16V - orange
4 = -16V - yellow
5 = +48V - green
6 = +24V - blue
7 = -24V - violet

 

[chunger]

well, I guess it couldn't all go without problems. . . and the problems I had were with these connectors.  

I think they were not designed to fit 16 gauge wire.  And, I probably shouldn't have been using 16 for this anyways (stiff and thick), but it's the one that I had in pretty colors.  Next time, I'll just cut a chunk of 8 conductor 20 awg off of my PSU wire and split up the strands and use those.

Back side of the connectors and I've changed the pin 1 grounding to chassis.

Front side

7 pin neutrik soldered up.  I tried the best I could to create clearance on the back side while using the silly 16awg wire that I will not do again in this spot.  I simply followed resistor color code numbering on the Neutrik pin positions.  I also did a little pigtail from pin 1 on the connector (chassis)  to the jack housing.  I hope that's proper.  I guess I'll have to look that up.

WARNING!  Do not use female connector for power input!  This could create a situation where there are exposed male power pins in the PSU cable and allow for accidental shorting of the main power feeds.  I will be changing this connector to male ASAP

I also bought a few of these little custom Neutrik plates from Redco (they can be ordered to say whatever you want).  These "spacers" will help create a little bit more clearance on the back side of the connector which is the primary reason I bought them.

Screws for the neutrik. . . the Neutrik goes on the plate, and I decided to stack the label plates to give me a little extra wire clearance without looking ugly.

So, it was all going to be down hill from here. . . simply screw in the wires to my terminals, button everything up, and we're done. . . First problem was I did not leave enough slack on the wire tails to maneuver easily. . .  next time, I should definitely leave another 2 inches or so to make things much easier. . . and then, on the 3rd wire, one of the screw terminal blocks stripped, and I was hosed.  The screw did not have enough threads to clamp down with a 16 gauge wire in to tighten down and the threads gave up.  So, I had to remove the connector.

I don't have proper de-soldering equipment, so these terminals didn't come out too clean.  Good thing these boards are built like tanks.  I was really hoping not to utilize that particular feature  

So, after this, I re-thought my strategy and decided to solder wire directly to PCB. . . The board has now been compromised so repair will be more difficult.  The front side of the holes have less damage than the back side of the PCB. . . I figure with 16 gauge wire, if I ever need to go back in for repairs, and really need to remove the Neutrik, I'll just cut and splice on the installed wire as that is pretty robust.

And, my final clearance is not too shaby so the spacer plan seems to have paid off.  

I soldered from the front side, and this is a shot from the back side. . . not too pretty, but I'm hoping it'll hold.  I poked a bit to make sure I had continuity and no weird shorts.



And, finally, assembly.

And, I had a kitteh in a box. .  .

Quite a few m2.5 screws go in the back side. . . (the back side screws are all the same even on the perimeter)

And, humans win!

 

[gar381]

Really nice job chunger !!

BTW.. love your cat !!  It looks like
it could be my cat's brother.

GARY

 

[Ptownkid]

Go chunger go!

 

[[silent:arts]]

I love the cat, I love the build, I love this thread.
need to read the post completely, but since I am in a rush:
never use female connectors for the PSU

 

[chunger]

OK. . . hadn't thought about that.  You don't want exposed pins on the live power feed leads so I'll have to change that to a male.  Crap, that was a hard one to solder together and expensive!

Live and learn. . . I've appended my previous post with a warning and will change this out as soon as I can.

 

[chunger]

Well. . . had a bad feeling that this wasn't going to end up as clean as I like. . . on my own, I would probably have left the female Neutrik, but since I know people are watching the thread, I figure I'd better make it right even if it ends up a bit messier than I like.  Guess I'll just have to buy another rack and do it right  ;D

Old connector is off.

At this point, I realized that the male connector has much smaller wire receiver lugs for soldering, so it would be absolutely impossible to get the 16 gauge wires in.  I had to go down to something smaller.  This time, I did what I should have done from the beginning (would have saved my block terminals also!). . . gut a piece of my 8 conductor power wire and use the 20 awg individual wires.

And, spliced in. . .

definitely not as clean as before, but solid I think.

and. . . after many errors and setbacks, humans win!

Let's try this shot again this time with the proper connector.

 

[Ptownkid]

Go humans! ;D

 

[jsteiger]

Go humans! ;D

One of my new favorite sayings is "humans win!!".

Thanks for that one 'Chung. 

 

[chunger]

Waiting now for PSU metal and perhaps parts.  In the meantime, I'm itching for something to test a few of the preamp modules with.  Probably pretty easy to figure out what I'm up to here:

I'm not intending for this to be able to power a whole rack by any stretch.  Just test a few modules at a time.  but after I get my official PSU built, I can hopefully put the little psu in the imaginary box pictured below to power 1 or 2 modules in a more portable fashion. . . that's the plan anyways.  Who knows if I'll be able to make it work.

 

[gar381]

Great Idea..

I know you will make it work chunger !!

GARY

 

[kepeb]

great stuff.
this is just whats needed for the 511 build

i like the little spacers for the power xlr, i think i must have been pretty lucky putting mine together i wouldn't fancy trying to get off that molex connector once it was on, had you soldered ALL the pins?

i did a similar thing with the wire to the backplane but with all the same colour :O

not sure if its because its the female plug you have on there, but the clearance didn't seem like a problem for me...
in fact, after i soldered the wires at tight right angles on the xlr, i was able to mount the socket inside.
this makes it much easier if you want to get at the back panel againfor any reason, as the metal comes off without having to disturb the wiring.

not sure about 'newbie' tho... i think your havin us on

 

[chunger]

Well,  Still no new word on the official PSU metal, so I'm tinkering away at my hack job solution.

The next problem I have been turning around in my head is how to mount the toroid flush with the L bracket cleanly.  I looked around to see if anyone made a dimple die that would work, and was surprised that nobody had anything like that, so it was back to old-school hammer-time for me. . . (car body type sheetmetal fab procedures).  I thought I'd post it up even though it's a bit off-topic because I haven't found a procedure like this on forum, and it would have been helpful to know.

First I prepped 2 scrap pieces of plywood to clamp my L bracket between.  One has a shallow 1" diameter hole, and the other has a 1 1/2" diameter hole all the way through.

Then, drilled a small hole in the center and I sandwiched the L bracket between with lots of clamps, found a somewhat suitable object (back side of a nail setter) and started pounding out a dimple.

And, the result. . . a bit messy on the finish with some hammer marks, but nothing that won't clean up if you're so inclined to sand it down.  I was not so inclined . . . also, this added a significant amount of rigidity to the L bracket due to the new shape as well as work hardening that happened to the aluminum as it was hammer formed.

I then cut down the bolt for the toroid by about 1/4" and did a test fit.

Not bad. . . just a little deeper than flush on the back side which is perfect.  This should slip into a single case without interference.

Next, it was just the usual drilling, hacking, and grinding to try and make the other bits fit.

And, here's where I stopped today. . .

The holes aren't perfectly aligned, and my component selection is let's say a bit clunky and fumbling still on anything that is not provided in a complete kit, but there's only one way to start figuring things out I guess.  I realize now that it'd probably be better to have used a 2 pole switch to switch both the neutral and the live. . . I had a fused power input module, but it would not fit the depth of both the L-bracket and the face plate.  Looking back, I could have easily cut the L bracket hole over sized and used that module.  Now, I'll have to put a separate fuse holder inside.  I'm hoping this will be ok with just the fuse on the mains because I don't have room to put 5 fuses for all the DC rails.

I skipped LED's for now just to get things done, and I it's going to get pretty tight inside with A/C and DC lines and toroids running way closer to each other than I would like already.

Also, I'm not quite sure how much this 30VA 12V x 12V avel transformer is going to be able to power and also on the same line, what value fuse to use on the mains.  I'm guessing a 1.5 or 2 amp fuse.  My primary concern is if I have enough heat sink for the 16v rails.  A lot of excess will get dumped into those regulators.  I figured the face plate was probably my biggest piece of available aluminum that could move some of that heat outside the rack.  The powerstation itself should be able to handle 3 amps, so that part should be ok. 

I wonder how many modules I will be able to power with this. . . I'm hoping 4 but will be able to use this if it can get 2.

 

[Holger]

Nice idea...
But I'm wondering if
- there is enough heat dissipation for the VRegs
- you'll notice any hum
If you are using a 30VA transformer (I'm not sure if a >30VA would fit) you will not be able to run modules in all 10 slots, my guess here.