Bipolar discrete shunt regulators

Status
Not open for further replies.
Jump to Latest
#222
hi cetoole,

Just a couple of points that "may" be worth considering. These are cosmetic and documentation factors only.

Centralise the output terminals.

Make OUT-1 equals IN-1 not IN-2 (if possible)

Delete extra nodes on C3 and C4 in the schematic.

7mm centres for resistor leads is a bit of a squeeze for "normal" 0.5 watt resistors.

Drill holes for C3 and C4 look big. ??

Have the inputs on the left hand side of the schematic and PCB, outputs on the right.

Number components left to right, top to bottom.

regards
 
#223
Well, Jan, Colin, Andrew and all of you reading this thread,
We seemed to be finished with this phase of the design. Do we need to build a new proto for testing purposes?

If yes, we would need some volunteer work to etch a board and populate it, anyone whishing to cooperate on this project please say so.
 
#224
Free Boards

Ok, let’s put this needed volunteer work in a more attractive perspective:

The guy who etches this double sided proto board will get for FREE a set of regulator boards from the GB when this is done. The number of boards will be determined by the Toolereg team at a later time and will depend also if the board is drilled and/or stuffed.

Please remember also that the GB will benefit DIYA with a small contribution as was set before.

Cheers 😎
 
#225
What value output voltage do we need for the LM4040? Just looking at the digikey list and they show several different parts (2.5V, 5v, 10v, 2.048, 4.096, 8.192 etc)

What values should the caps be? Can we use 220uF for C1 and 8.2uF for C2 (or 4.7uF or 10uF). How critical is it to use poly for C2, since I see Colin has put the option of using a pair of electrolytics instead. I have some Elna silks @ 4.7uF, would those work better than a 10uF panasonic pp?
 
#227
twitchie said:
What value output voltage do we need for the LM4040? Just looking at the digikey list and they show several different parts (2.5V, 5v, 10v, 2.048, 4.096, 8.192 etc)

What values should the caps be? Can we use 220uF for C1 and 8.2uF for C2 (or 4.7uF or 10uF). How critical is it to use poly for C2, since I see Colin has put the option of using a pair of electrolytics instead. I have some Elna silks @ 4.7uF, would those work better than a 10uF panasonic pp?
Click to expand...


The only requirement for C2 is that it is lossy, so NO 'high-quality' boutique caps with funny parasitic inductances and very low ESR! Any electrolytic will be fine. Lowest cost.

Jan Didden
 
#228
I've tried to revise the layout so I can etch it single-sided. I had to incorporate 3 jumpers. Could someone let me know if I've compromised anything in the layout?

I tied R1 gnd to R8, IC1 and C1 gnd and changed the footprint for C2 to 2mm lead, 5mm grid (for the 10uF electolytic I have).

An externally hosted image should be here but it was not working when we last tested it.
 
#229
Twitchie,

That’s a very nice effort from my point of view, I see that C2 is quite small but seems enough for a 100u @25V if possible I would add an alternate say a 10mm dia with 5mm leadspace. You could rotate LED3 and move it downwards to make space for a bigger C2.

We will wait and see what others have to say about this one sided alternative.

Anyway you are only runner up this far 😉
 
#232
I would probably do some things different, but that's more a matter of personal preference and not performance.
What you can try is to move C2 down, D1 and R7 up and to the left and RF2 up, to make space to move the input connector inwards. That'll make your board smaller & cheaper to produce and takes less space.

Jan Didden
 
#233
Compacted a bit further but in my opinion, it's reaching the point of diminishing returns.

I etched the original pre-Jan revision and the new version is already smaller than that one. If someone wants to etch this thing I'll upload pdfs as soon as I get a second pair of eyes on it letting me know there aren't any mistakes.

It doesn't mean I won't etch it, but I'm about as fast as GM's economic recovery :xeye: when it comes to finishing things :cannotbe:

besides, I am still missing LM334 and LM4040 (going to order 5V variant unless I hear otherwise) and I wouldn't know where to begin testing this thing.

Enjoy:

3335747941_b8506ba39e_o.jpg


It's pretty small in reality:

3336583150_5032a021be.jpg
 
#234
cooling

Hello,

is some cooling needed for Q1 and other TO220 transistors. If yes, how did you guys preview to do that since the transistors are quite "inside" the pcb?

Regards,
Pred

PS
I was thinking to pass like 100-200 mA trough this reg 🙂
 
#237
Is the current cct design as per post 220 is the final version?  [with the LM4040s, LM334z, no CCS option for R5, the unknown 1uF (poly?) caps for C3, C4, and a no-name o/p electrolytic cap]  And will we be using this component numbering in the future as it's all "jumbled" now.
The variable resistor (VR15/ VR1/Pot) )has been excluded - why?Twitchie, on your layout (post233) move link from R1 to the O/P terminal I still see at least 5 seperate earth tracks to central 0V point on the o/p pin
 
#238
Sorry for the rough format in previous post - missed keystroke!

Is there a matching series regulator and/or a method to reduce ripple and line noise to precede this shunt?

Nothing wrong with links, twitchie. Suggest increase width of tracks tho. amd reduce lengths where possible.
 
#240
Ok, so curiousity got the better of me and I went ahead and etched, drilled and stuffed this thing, but the leds don't light up and it just drops about 0.47V off of the input voltage. It just follows input voltage all the way through the 27V or so I tested it up to. I was expecting it to start shunting voltage and hold at 18V or so, but that wasn't the case.

At first I thought maybe the LM4040 were supposed to be 10V variant and swapped in 9.1V zeners instead, but no change whatsoever.

I'm feeding it from a LM317 psu and used the exact resistor values in Colin's schematic. leds are not in backwards and C1 is 220uF and C2 is 100uF
 
Status
Not open for further replies.
Top Bottom