Power supply using a 2U rack chassis (56K warning)

A place to document your own projects.

Re: Power supply using a 2U rack chassis (56K warning)

Postby hak8or » Thu Sep 08, 2011 12:27 pm

bearmos wrote:back when i was using ISE Web Pack (not sure if that's still the name) the schematics were viewed by my professor as a cute feature, rather than something useful! A lot of times, we would use an external flow charting tool to quickly sketch what we were making, rather than get bogged down in the tedium of schematic entry.

If you have a solid digital logic background the VHDL is a bit cumbersome, but no more than any new language - mainly due to syntax.

Then, eventually, you'll get sick of writing the repetitive VHDL (if you're doing purely structural, building everything from NAND gates) - that's where the hacked together VHDL generators come in - and the fun really begins!

I need to get back into this, it's been too long!



ian wrote:I always have problems flipping and moving parts, and it often won't autoroute to somewhere due to some clearance you can't see at all. The frustration experienced with the cct entry is a well known bug ;) You would not be alone in hating it. I used it for a long time and eventually even learning VHDL/Verilog was less painful (not that I would encourage you to do that, it was a year with schematics before I was ready to try code).


I am planing on learning VHDL or Verilong some time soon, hopefully within the year. The schematic entry is as your professor would put it, a fancy toy. I want to be able to get familiar with as much of the ISE as possible, including the schematic entry, and I know how to do things with visual logic instead of the language alright, so I might as well start with something that I am already able to reasonably control. I would rather learn the CPLD first through finding out what type of logic function is optimized in what way, things like that, first and then start working on learning another language. In other words, I want to get familiar with the platform first, and then start to work with it in a more controlled/complicated manner, even if that more complicated and controlled manner may be faster.

I am eager to start to learn how to code on the CPLD and FPGA, but if I do it that way I will end this power supply project in months and months instead of just weeks, heh. :P I do have some experience with FPGA's and their languages when I used the Logic sniffer as a development board. As a matter of fact, I bought the logic sniffer mainly for using it to learn about FPGA's (Development Board), while the logic sniffing part of it was not that important to me. I also bought it as a thanks to the community for making it.

I also found that I really enjoy working on this using the schematic entry. For some reason, I love to see the busses in there. I guess the reason for that is when I read the book called "microprocessors" which is about the "ancient" 8088 processors, they sometimes showed the buses, the thinking was done in terms of buses which were very clearly marked compared to individual lines. Having read that just a year ago, and loving the idea of parallel data lines, I just had to incorporate that into my design.

The external flow chart program that you mentioned before sounds very interesting, and it would surely be awesome to see how your verilog/VHDL code looks in a flow chart view. From what I remember, I think the Xilinx ISE has a code to schematic tool. I am not sure though, as I have never used it so I can't even state how useful it is, but I should have a look at that when I start writing FPGA and CPLD things in code.

Oh dear, this post is rather long, as are most of mine, I think I should shorten them. :P Any opinions on the project so far? Does it look like a waste of time, does it look awesome, is anyone other than Ian and Bearmos reading this? Advice is always welcome too! :)
hak8or
Full Member
Full Member
 
Posts: 241
Joined: Mon Jun 06, 2011 11:06 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby bearmos » Thu Sep 08, 2011 1:06 pm

hak8or wrote: In other words, I want to get familiar with the platform first

I understand completely, project scope has to be drawn somewhere:)

hak8or wrote:The external flow chart program that you mentioned before sounds very interesting, and it would surely be awesome to see how your verilog/VHDL code looks in a flow chart view

This was really just MS Visio (or Open Office Draw) - just something to quickly generate block diagrams to keep track of high-level ideas. There was nothing fancy going on between VHDL and the diagram - it was just easier to email a visio block diagram than a napkin:)
bearmos
Hero Member
Hero Member
 
Posts: 748
Joined: Tue Nov 02, 2010 10:02 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby rsdio » Thu Sep 08, 2011 10:36 pm

hak8or wrote:The xilinx tools do allow for the creation of a bus, but I do not want to use it in this case because .. well ... wait, why am I not using a bus?
Ha! These are the kinds of questions I ask myself all the time.

Oh, yes. I had to open the xilinx schematic entry window and look at my schematic again. It simply helps me see what I am doing. It does make sense to be using a bus when doing something like this, but I want to have each of the connections visible to me within reason to aid in understanding what I am doing.

Eagle allows you to View a signal and it will be highlighted on every copy of the bus as well as on the pins of each connected chip. That makes it easy for me, at least, to see what's going on in my Eagle schematics. I haven't used the Xilinx tools either, though, so I don't know whether they have something similarly useful.
User avatar
rsdio
Developer
Developer
 
Posts: 1410
Joined: Sun Feb 28, 2010 10:53 pm
Location: Seattle

Re: Power supply using a 2U rack chassis (56K warning)

Postby hak8or » Fri Sep 09, 2011 1:57 pm

rsdio wrote:
hak8or wrote:The xilinx tools do allow for the creation of a bus, but I do not want to use it in this case because .. well ... wait, why am I not using a bus?
Ha! These are the kinds of questions I ask myself all the time.

Oh, yes. I had to open the xilinx schematic entry window and look at my schematic again. It simply helps me see what I am doing. It does make sense to be using a bus when doing something like this, but I want to have each of the connections visible to me within reason to aid in understanding what I am doing.

Eagle allows you to View a signal and it will be highlighted on every copy of the bus as well as on the pins of each connected chip. That makes it easy for me, at least, to see what's going on in my Eagle schematics. I haven't used the Xilinx tools either, though, so I don't know whether they have something similarly useful.


I am not sure if xilinx schematic view has such a feature. It most likely does somewhere where I have not seen it yet.

bearmos wrote:This was really just MS Visio (or Open Office Draw) - just something to quickly generate block diagrams to keep track of high-level ideas. There was nothing fancy going on between VHDL and the diagram - it was just easier to email a visio block diagram than a napkin:)


Heh, now I understand what you meant. :P
What very often happened to me in high school was I was sitting in class, bored out of my mind since my teacher was explaining the same thing the third time to the students, and I drew designs on what I had at the time. So, my notebook is half notes and work, the other half was ideas that I drew, the tests I got back had schematics on them while the teacher was reviewing the test results with the students, and worst of all is when the teacher does a random note collection. Imagine the teachers confusion when I had the drawings in Chemistry slowly changing into schematics. :P


A quick thing I wanted to say, this is a very enjoyable forum! The people here are all very nice and energetic, and I am slowly starting to remember all your usernames, which has never happened on any other forum. One thing for sure, I will definitely post my project write ups on here from now on! And, hopefully within the next two years I will present something which I have been wanting to do for years, since 8th grade, which was nearly six years ago.

Quick question: Are there any replacements for a tri state buffer for Xilinx fpga's in the schematic view? Or, is it possible to have a high impedance symbol? I found that multiplexers make ISE happy but from what I understand a pin when not connected to anything gets connected to ground instead of left in a high impedance state.
hak8or
Full Member
Full Member
 
Posts: 241
Joined: Mon Jun 06, 2011 11:06 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby bearmos » Fri Sep 09, 2011 2:24 pm

hak8or wrote:A quick thing I wanted to say, this is a very enjoyable forum! The people here are all very nice and energetic

I agree, I'm continually surprised by the caliber of the members here.
bearmos
Hero Member
Hero Member
 
Posts: 748
Joined: Tue Nov 02, 2010 10:02 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby hak8or » Wed Sep 14, 2011 10:28 am

Just wanted to let you guys know, this project is still alive!
I am working on learning verilog, and working on this project with what ever little time I have. I am considering leaving the FPGA design out for now (Entire high speed ADC will be left for later) and instead working on the other parts some more. The PCB will have the pads and traces for the FPGA, Memory, and ADC all on it, and the chips soldered too, but I will just program the FPGA later. :P

I also might do a simple design, like the bus pirate, as a test for my PCB skills. I nicked some large scanner from my old schools garbage on garbage day, so I think it will make a fine UV exposure box. Just need to get a photo resist developer first.
hak8or
Full Member
Full Member
 
Posts: 241
Joined: Mon Jun 06, 2011 11:06 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby hak8or » Wed Sep 21, 2011 11:20 am

Still not dead, I am learning verilog in the free time I have at college between classes.
Verilog does indeed seem much faster and less cumbersome than schematic entry! Always good to learn another language. :P
hak8or
Full Member
Full Member
 
Posts: 241
Joined: Mon Jun 06, 2011 11:06 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby hak8or » Thu Sep 29, 2011 1:30 pm

Hello everyone again!

I have had major internet problems for the past week and half or so, so I have been working with Time Warner Cable to get my connection back. In the meantime, I was able to spend more time on the project which resulted in almost finishing the PCB and schematic. I will upload soon, if I can upload it at all with this connection.

I needed to use a more "High end" suite for the board since I have a few 50 MHZ data lines, in which case I did trace length tuning. Thanks to all this time and thinking I got some more ideas, for example, I want to be able to test reflection in high speed traces. I can use a FPGA, a few comparators to see if a few specific voltages appear on the line, and a counter to find the gaps between sending and a reflection. The actual term for this seems to be a Time Domain Reflectometer.

Image
http://archive.hak8or.com/PSU_BIG/Schem_1.png

Image
http://archive.hak8or.com/PSU_BIG/PCB_1.png

http://archive.hak8or.com/PSU_BIG/PCB_Project4.pdf <-- PDF
Last edited by hak8or on Mon Apr 06, 2015 10:04 pm, edited 1 time in total.
hak8or
Full Member
Full Member
 
Posts: 241
Joined: Mon Jun 06, 2011 11:06 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby hak8or » Wed Oct 05, 2011 12:12 am

I am almost done with the board, just need to put in some last touches. Finished with the "expansion" side of things. I will be using a high speed SPI (40+ MHZ) from the fpga, which can be later plugged into another FPGA board. The other FPGA board will act as an I/O manager, basically the controller for additional peripherals I add in later. The same SPI connector will act as the bus between the propeller and the FPGA. The propeller will be used as the GPU mostly, and the FPGA holds the information, with the Prop fetching data from registers in the FPGA. I am using a additional connection (SPI has three) for addressing between the prop and the add on FPGA board.

Off topic, but I wanted to show you guys what I have been playing around with a bit lately, reverse lense photography. Basically, get a lens for DSLR's and shoot pictures through it with your camera. It gives you a not bad macro ability without purchasing a macro lens.

http://imgur.com/a/8WMYa#9k01n
First three are from a GPU (X800, first graphic card I had :D )
Anyone want to guess what the last two pictures are?
hak8or
Full Member
Full Member
 
Posts: 241
Joined: Mon Jun 06, 2011 11:06 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby ian » Thu Oct 06, 2011 6:34 am

It gives you a not bad macro ability without purchasing a macro lens.


Very cool!
User avatar
ian
Crew
Crew
 
Posts: 10803
Joined: Mon Jul 06, 2009 6:14 am

Re: Power supply using a 2U rack chassis (56K warning)

Postby hak8or » Wed Nov 09, 2011 12:21 am

Very annoying problem for me right now, I purchased all the chemicals for making photoresist boards, and I was excited about making a test PCB. I do a test to see how small traces my printer can handle, and it turns out that the smallest traces I can make are only maybe 8 or 10 mil! This is extremely upsetting! Not to mention that the quality of the traces is rubbish to say the least. I would have to make maybe three copies and put them onto of each other to get a good light blocking.

The main reason I am unhappy with my situation is because I purchased this printer a few years ago under the idea that it is a fantastic printer, with a decent DPI of 1600 which would give me one dot per .625 mil. I believe that thanks to this I should be able to get 6 mil without too much trouble, but I guess not. Maybe if I run over the traces with a fine pen I might be able to do better. Also, I am still interested in my idea of using a camera lens I have to make the transparency half its size on the pcb. Not to mention, I became very curious about using a metal heated head to "print" wax onto copper boards and etch that way, thanks to this presentation a while ago.http://hackaday.com/2011/01/31/printable-wax-as-pcb-etch-resist/

Also, for some very odd reason, I have started to get interested in the photo lithography process for making ic's, heh.
Woe is me, for all these distractions end up with many unfinished projects! :P
hak8or
Full Member
Full Member
 
Posts: 241
Joined: Mon Jun 06, 2011 11:06 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby ian » Wed Nov 09, 2011 5:28 am

Assuming inkjet: Be sure to use "high quality black" with a mix of colors to make black, yellow is usually most UV resistant. Get photo ink, which is UV resistant. TUrn up the ink volume to make a thicker mask, but be sure to let the transparency dry or it will get hair and dust in it. Be sure to print backwards so the ink is against the PCB for best results. Be sure your light source is direct and is close to the PCB. Also, practice makes perfect ;) I did everything at 24 and 32mil for a long time before I could etch QFN by hand.
User avatar
ian
Crew
Crew
 
Posts: 10803
Joined: Mon Jul 06, 2009 6:14 am

Re: Power supply using a 2U rack chassis (56K warning)

Postby bearmos » Wed Nov 09, 2011 8:30 am

hak8or wrote:Not to mention, I became very curious about using a metal heated head to "print" wax onto copper boards and etch that way, thanks to this presentation a while ago.http://hackaday.com/2011/01/31/printabl ... ch-resist/


I wouldn't even go down that path if I were you, based on what I've seen its a very experimental rabbit hole. If you'd like to do it as a project, it's a different story though. It's all too easy to get swept off into the side projects (I'm speaking from first hand experience here ;-)). At the end of the day, if you enjoy what you're doing that's what matters though!

I've never used the photo resist method (i've always ironed on laser printed patterns). The finest traces i was able to produce were 16 mils and even that was very painful. Part of the issue I have was that the printer needs a new fuser, so the quality is pretty low. Eventually I was able to etch a 64 pin QFP, though.

Home etching definitely requires plenty of experimentation/patience. I've been trying to get a home-brew mill setup for a while with limited success. The idea behind "mechanical etching" is that you can theoretically panel plate the PCB before milling the traces (giving a plated through hole board). That way you don't need to worry too much about putting vias under components, threading wire through vias and soldering, etc. Again, this type of thing starts as a side project and then turns into a rabbit hole/time sink.
bearmos
Hero Member
Hero Member
 
Posts: 748
Joined: Tue Nov 02, 2010 10:02 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby hak8or » Wed Nov 09, 2011 5:48 pm

bearmos wrote:
hak8or wrote:Not to mention, I became very curious about using a metal heated head to "print" wax onto copper boards and etch that way, thanks to this presentation a while ago.http://hackaday.com/2011/01/31/printabl ... ch-resist/


I wouldn't even go down that path if I were you, based on what I've seen its a very experimental rabbit hole. If you'd like to do it as a project, it's a different story though. It's all too easy to get swept off into the side projects (I'm speaking from first hand experience here ;-)). At the end of the day, if you enjoy what you're doing that's what matters though!

I've never used the photo resist method (i've always ironed on laser printed patterns). The finest traces i was able to produce were 16 mils and even that was very painful. Part of the issue I have was that the printer needs a new fuser, so the quality is pretty low. Eventually I was able to etch a 64 pin QFP, though.

Home etching definitely requires plenty of experimentation/patience. I've been trying to get a home-brew mill setup for a while with limited success. The idea behind "mechanical etching" is that you can theoretically panel plate the PCB before milling the traces (giving a plated through hole board). That way you don't need to worry too much about putting vias under components, threading wire through vias and soldering, etc. Again, this type of thing starts as a side project and then turns into a rabbit hole/time sink.


Thanks for the information Ian and Bearmos! :P And most certainly it would be a project if I were to do it! :P It seems very interesting, but I would certainly try to make my own head, problem is making such super tiny holes. Maybe I can get someone somewhere to fabricate it for me with super super tiny holes but made of some sort of metal. From there on I could probably work on it myself such as the electronics and thing to push the ink out as well as keep the head warm. The CNCish things though might be a bit complicated, I was never that into the machining world, so yeah. :(

Also, yeah, heh, if it has any sort of gain (enjoyment and fun, some application for what you do, or just to learn something) it is fully worth it, in my opinion at least.

Also, Ian, thanks for all the information, but my printer (Lexmark crap) does not have that much control. :( Not to mention the ink is very expensive, especially the photo ink. Right now I am kind of leaving the lexmark printer idea behind and instead looking at making my own printer thing, using wax or just using a lens from designs 2x to smaller, which decreases the tolerances needed for my printer. Problem is that I have a UV filter stuck on the lens, heh. I dropped the lens a long time ago and the UV filter is stuck on the lens, have to get it off somehow. Was thinking about heating the metal rim of the UV filter hoping it expands enough to nudge it out of place, but I am not sure if the glass or surrounding material would be happy about that. :(

Anyways, thanks for all the suggestions guys! And thanks for the front page again Ian! :D
hak8or
Full Member
Full Member
 
Posts: 241
Joined: Mon Jun 06, 2011 11:06 pm

Re: Power supply using a 2U rack chassis (56K warning)

Postby Sjaak » Thu Nov 10, 2011 2:09 am

You could use a needle from a syringe, which comes in lots of different sizes. The problem would be how to squeeze and control a supersmall droplet from it.
User avatar
Sjaak
Fellow
Fellow
 
Posts: 3297
Joined: Sun Jan 03, 2010 2:45 pm
Location: Hiero

PreviousNext

Return to Project logs