Jump to content

Midibox PSU


NorthernLightX

Recommended Posts

I wasn't using your design but a design of my own. Just sharing a lesson I learned the hard way.  ;D

Oh I knew you didn't use my design, that would be quite hard since it's only been up for a day or so :) . I meant that the partlist will include overrated stuff to prevent faults like you had.

Anyone who want's to build this PSU should note carefuly, that polarization IS an issue if you also decide to not stuff the rectifier and stuff on your boards. It might be a good idea to color code the connectors in some way ;)

Maybe the connectors could be a little easier to get to....

I'm not going to change the design if there are no technical reasons. the connectors are easy enough to reach as long as you make use of female SIL plugs for the connectors (only the big elco's are "high" components). Soldering directly to the headers might be a bit of a problem, but I do not recommend that anyway.

Cheers,  Alex.

Link to comment
Share on other sites

  • Replies 108
  • Created
  • Last Reply

Top Posters In This Topic

One last thing, about the rating of the big resistors. For the not so techical readers (like me!): The theory behind them is as follows: They will dissipate an amount of Volts equal to the amount of Ohm they're rated for, at a current of 1 Ampere. If the current is less than 1A they will dissipate less, but the regulators will be stressed less too, so this is exactly what is wanted.

I'm not 100% sure if the rating of the resistors is good in the current design. I did some quick calculating after learing about the above rule yesterday:

The power out of the mains can be plus or minus 10% of the specification (115V or 230V). Let's assume its -10%, as it is in my house.

So, instead of the 12V the trafo delivers at 230V, it in fact delivers 10,8V (12V - 10%).

When rectified and filtered, the voltage becomes 10,8V * 1,41 = about 15,23V.

this is perfect for the 7812 and 7912, so far so good.

If we assume (for simplicity's sake) that there's a load of 1A behind the 7805, the resistor R1, which is rated for 5 Ohm, will dissipate 5 Volts. But since the voltage after the rectifier is 15,23V, there will still be 10,23 V left, a tad high for my taste. At the 7809 it's the same story, 15,23V - 2,5V is still 12,73V. The difference here is a little less disturbing, but note that these numbers are the best case scenario!

If for example the mains is not -10%, but +10%, the voltage before the 7812 and 7912 is 12V * 1,1 * 1,41 = 18,61V. Not something to worry about if you got a decent heatsink on them. But for the 7809 an 7805 it's a different story, they will be fed with 18,61V - 2,5V = 16,11V and 18,61V - 5V = 13,61V. I find that much too high.

I could be paranoid and/or have misunderstood the calculations to take, so if anyone sees flaws in my logic please reply! :)

My suggestion would be to replace the 5 Ohm resistor for 8 Ohm, and the 2,5 Ohm resistor for 4 Ohm. That way the calculations would be as follows:

voltage @ 7805 @ -10% = 15,23V - 8 = 7,23V

voltage @ 7805 @ +10% = 18,61V - 8 = 10,61V

voltage @ 7809 @ -10% = 15,23V - 4 = 11,23V

voltage @ 7809 @ +10% = 18,61V - 4 = 14,61V

These are values anyone should be able to live with :)

As usual, comments are welcome!

Cheers,  Alex.

Link to comment
Share on other sites

I could be paranoid and/or have misunderstood the calculations to take, so if anyone sees flaws in my logic please reply! :)

Heheheh well, if your logic were flawed, I'd have no clue, but I don't think you're paranoid... The juice at my place tends to run pretty high, like 250v all the time, and even though everything runs off UPS, I still like to know that I can cut it out and not have all my power supplies simultaneously explode ;)

Link to comment
Share on other sites

Indeed, the only component you need to change is the transformer, exchange it for a specimen that converts 115Vac into 2x12Vac. After that, all components are the same.

Please do note that another change to the design is being made: if an 8 Ohm resistor is going to be used, it also has to be at least 8 Watt. So, you only have to swap it then, right? Wrong, this baby is even bigger, and I have to create a new board (again) because the stuff doesn't fit anymore  >:(

Almost finished, will post it tonight.

Link to comment
Share on other sites

[update]

Alle images were moved to the WIKI: http://www.midibox.org/dokuwiki/northernlightx

[/update]

As the board had to be slightly bigger to accomodate the 9W resistor, I reserved a little (and i mean a little but it could have been less ;) ) space around the SIL headers for easier access, as per stryd_one's request.

Preliminary orderlist @ Reichelt:

+----------------------------------------------+-------+--------------------+-------+--------+
| Description                                  | Qty   | Part #             | Price | Total  |
+----------------------------------------------+-------+--------------------+-------+--------+
| Voltage Regulator 7805                       |     1 | uA 7805            |  0.17 |   0.17 |
| Voltage Regulator 7809                       |     1 | uA 7809            |  0.17 |   0.17 |
| Voltage Regulator 7812                       |     1 | uA 7812            |  0.17 |   0.17 |
| Voltage Regulator 7912                       |     1 | uA 7912            |  0.17 |   0.17 |
| Resistor 220 Ohm                             |     1 | 1/4W 220           |  0.10 |   0.10 |
| Resistor 4 Ohm  5W                           |     1 | 5W AXIAL 3,9       |  0.29 |   0.29 |
| Resistor 8 Ohm 9W                            |     1 | 9W AXIAL 8,2       |  0.36 |   0.36 |
| Polarised Capacitor 4700 uF                  |     1 | rad 4700/35        |  1.00 |   1.00 |
| Polarised Capacitor 2200 uF                  |     1 | rad 2200/35        |  0.46 |   0.46 |
| Polarised Capacitor 22 uF                    |     4 | rad 22/35          |  0.06 |   0.24 |
| Ceramic Cap 220 nF                           |     2 | Z5U-2,5 220n       |  0.13 |   0.26 |
| Ceramic Cap 100 nF                           |     4 | Z5U-2,5 100n       |  0.06 |   0.24 |
| Diode 1N4008                                 |     4 | 1N 4008            |  0.02 |   0.08 |
| 1-row SIL Headers (about 14 pins)            |     1 | STIFTL. 40G        |  0.18 |   0.18 |
| Ringkern/Toroid Trafo 2x12V 50VA             |     1 | RKT 5012           | 12.30 |  12.30 |
+----------------------------------------------+-------+--------------------+-------+--------+
| Total                                                                                16.10 |
+--------------------------------------------------------------------------------------------+

Note 1: Heatsinking of the 78xx regulators is a must. It's possible to create a heatsink from a  
        piece of aluminium that you cold attach to all 3 regulators.
Note 2: The 7912 regulator has a different pinout than the 7812, the metal slab is 12V in, 
        so take the necessary precautions if you want to heatsink it!

Optional parts (not necessary but very handy!)
+----------------------------------------------+-------+--------------------+-------+--------+
| Description                                  | Qty   | Part #             | Price | Total  |
+----------------------------------------------+-------+--------------------+-------+--------+
| Female Power plug                            |     1 | KES 2              |  0.69 |   0.69 |
| Fuse holder                                  |     1 | HALTER 10,3X38     |  2.10 |   2.10 |
| OR Combined  Power plug + Fuse               |     1 | KES 1SI            |  1.35 |   1.35 |
| Power switch                                 |     1 | DS 059 RT          |  1.55 |   1.55 |
| 1-row FEMALE SIL Headers                     |     1 | BL 1X20G8 2,54     |  0.34 |   0.34 |
+----------------------------------------------+-------+--------------------+-------+--------+

As usual: any comments are welcome. I really hope this was the last design change ;)

Cheers,  Alex.

[edit]

Last minute change: the elco on the -12v rail has been replaced with a 2200uF specimen. Saves you another 54 cents ;) (and NO changes to the layout for a change! :D )

Also I added another image, of the copper tracks without the components.

[/edit]

Link to comment
Share on other sites

I finally found out how the polygon option works :)

Pics have changed slightly, and by replacing one of the 4700uF elco's with a 2200uF, another 54 cents can be saved  :P

I've placed an order at Reichelt for the parts, and Mendelt, Wisefire, and myself are going to betatest the design. PCB's will be ready in 3 weeks (not for resale, sorry, but when the design is tested and found ok I will release the schematic and the .BRD to the community so no worries ;) )

Chees, Alex.

Link to comment
Share on other sites

Would a 2x12 30VA rated transformer be sufficient? I noticed you had a 50VA rated one on your list...

That depends on what you're planning to run on it. If your total load on the positive branch does not exceed 1,25A it is sufficient:

With 50VA, you have 25VA per branch (in my design, this will be the positive and negative branches). 25VA at 12V means a max load of 2,0833 Ampere behind 7812, 7809, and 7805 combined. The 7912 on the other hand has the total of 2,0833 A for itself.

With 30VA, you have 15VA per branch. 15VA at 12V means a max load per branch of 1,25A.

Note that each 78xx can handle a max current of 1A, it is not possible to exceed that. So for example, if your setup draws close to 1A on the 5V branch, with a 50VA trafo you still have more than 1A left for the combined 9V and 12V branches. With a 30VA trafo your headroom in this situation is only about 0,25A.

I hope this answers your question ;)

Cheers,  Alex.

Link to comment
Share on other sites

I thought you said you were new to this! :D

Hey, do something right or don't do it at all is the motto!  8)

I am very new to this, but much reading and discussing stuff with people who do have the knowledge helps a lot. ;)

Besides, this might becoma FAQ stuff, and if the proper documentation can already be found in this thread that leaves a job of only cutting and pasting when the testing phase is completed.

Now, if only those biatches at Reichelt would be a little quicker at answering my email.....

Link to comment
Share on other sites

Will you be distributing the BRD file for this nice piece of work?

Absolutely, but I think it would be better to test it first. Parts have been orderen from Reichelt (and they don't respond to email anymore as soon as you ask why the price is for a foreign order is 50 euro more than a domestic one  >:( ) and also PCB's are arranged for. If the results are as we expect, .BRD and >SCH files will be released. (Into the WIKI I suppose).

I -might- order a few extra PCB's if there's demand for it, when the PSU had been approved.

Cheers,  Alex.

Link to comment
Share on other sites

Glad to hear. Hope everything goes ok with Reichelt. By the way, upon looking at your schematic, I noticed you erroneously have two 4700 caps instead of one 4700 and one 2200. Had me confused for a second there! No doubt you probably already fixed it, but just wanted to let you know what is on the thread.  ;)

Link to comment
Share on other sites

  • 2 months later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...