Voltage display 25V?

[Lliam20789]

Hello everyone,

What is the best way to monitor a Home-Made DC 22V Max Variable Power Supply. I would like to have the voltage displayed on a few (maybe 3) 7 segment displays if its not to hard, any ideas or suggestions?

Thanks,
Lliam

 

[BeanieBots]

Define "best".
Cheapest, easiest, most accurate?
If all of the above, then digital panel meter?
If as PICAXE project, then /5 potential divider and readADC10.
That will give about 0.1% resolution which should be possible to converted to 0.1% accuracy after calibration and correct choice of supply regulator.
Remember, the accuracy of the A2D is only as good as the accuracy/stability of your power supply.
Once you have that part done, there are many methods to display the result.
Multiplexed 7-segment display, 4026 counter driver chips. AXE033 LCD, barebones LCD. Many examples can be found in the archives.

 

[Lliam20789]

Yeah, all the above for "best".

Hopefully salvaging most parts needed.
Resolution is not of great concern, within 0.1 of a volt would be a good minimum.
Just for the record, its a modified computer power supply so fairly reliable.
What do you suggest for the display? I'm hoping to get away with 08-M's only.

Thanks

 

[BeanieBots]

Resolution is not of great concern, within 0.1 of a volt would be a good minimum.
Just for the record, its a modified computer power supply so fairly reliable.
What do you suggest for the display? I'm hoping to get away with 08-M's only.
Thanks

Should be possible to get about 0.02v resolution.
Modified PC power supply is NOT a good choice. You want a good stable ideally linear power supply for analog work. What else will that power supply be driving? Will it be correctly loaded? Any noise on the power line will affect the readings.
08 does not have the required ADC resolution. (about 1.5v resolution)
08M has limited outputs so eliminates the possibility of multiplexed 7 segment or barebones LCD without extra hardware.
AXE033 would be the easiest.
3 off 4026+7-seg display would be simple/cheap but a lot of connections to make.
Choice is yours. Only you know your strengths & weaknesses.

 

[Michael 2727]

You can use 2 pins to drive a 4026 driven 7-Seg display.
A Pulse for the Count and a Reset are all that are needed.
Try this thread -
http://www.picaxeforum.co.uk/showthread.php?t=7764

Getting the ADC resolution needed may be the tricky part
as BeaineBots has pointed out.

 

[Lliam20789]

Interesting,
The power supply was going to be used on small picaxe projects/tests...
I was going to have a load resistor, not sure of value as of yet.
I guess I could use an 18A instead of 08M's if required.

Do you have a better suggestion for a power supply?
It would be good if I could make it mainly from salvaged parts.

 

[Michael 2727]

If you want a variable supply use a transformer, bridge rectifier,
filter capacitors (larger the better) and an LMXXX 3 Pin variable
regulator, on a heat sink (larger the better).
They are available in 1A, 3A and 5A versions.
The 3 Pin Reg can handle voltages up to 40V, although keeping
the input MAX voltage close to the MAX output voltage you
need will let the Reg run much cooler, e.g. 24V input to 20V output.
( a 15V to 20V Transformer, higher current rated the better, will do
for most hobbyist apps )

And most of this and sometimes including fixed V Regs can
be scrounged/recycled from old VCRs TVs Stereos etc.

 

[Lliam20789]

Thanks, that sounds like the go.
I'll have a look for parts at school, they have a few dead bench power supplies...
So a supply like that could easily run the picaxe monitoring stuff as well as 2 separate variable supplies?
What would be the best way to power the Picaxe system, as large voltage drops would cause lots of heat?
Should I incorporate a fuse into this design as well?

Thanks for you guidance.

 

[Dippy]

Good luck Lliam and yes fuse it.

To save embarrassment/hassle I don't suppose you can find a s/hand good bench PSU which has current limiting too?
This facility has saved me many re-orderings / tantrums / grey hairs etc.

I would like the following:
5V fixed output with current limiter
0 - 30 V 5A with current limiter
DIgital displays for V.Voltage and current limit.
Under a tenner. Not made in China.

My TTi is great but sadly not cheap.

 

[Lliam20789]

Nope, looked on ebay, no such thing and haven't got anywhere else really...
Would it really be a difficult task to make it from scratch?
If not, would two transformers or two regulators be the best option?

I agree with you Dippy that a fixed 5V and variable ~30V, Few Amps with display, will save parts and still do everything.

 

[Mycroft2152]

A couple of thoughts on your power supply. If you are building one for the sake of learning about power supplies and developing building skills, great. But on a practical and cheaper basis:

Use an old cell phone charger, they usually provide 5 volts at 1 amp in a very small package.

Use a cheap chinese digital meter for the voltage and pmerage displays. They can be purchased for about the same price as a PICAXE 08M chip ($3US).

Then use the PICAXE as the current sensor limiter.

Myc

 

[Dippy]

Yes, sounds good . Or maybe a single transformer with dual/multi outputs?
Your rectifiers / regulators will have to be hefty types and some quality capacitors.
You could even have a PICAXE or op-amp controlled over-current trip, which I would say is almost obligatory when prototyping esp hairy bread-boards.
You'll need a meter (or hacked mulitmeter) to monitor variable volatge o/p.
And a nice tough box if you want it nice, neat and safe.
Price is starting to mount....

Will it be cheaper than this:
http://www.rapidonline.com/productinfo.aspx?tier1=Electrical+&+Power&tier2=Power+Supplies&tier3=Bench+PSUs&tier4=0-30V+Triple+output+analogue++DC+regulated+PSU&moduleno=77151
OR..
http://www.rapidonline.com/productinfo.aspx?tier1=Electrical+&+Power&tier2=Power+Supplies&tier3=Bench+PSUs&tier4=18V+Dual+LCD+meter+PSU&moduleno=74610
(no doubt made in PRC)

It'll certainly be cheaper than this:
http://cpc.farnell.com/jsp/Test+Equipment/Electrical/Electronic/TTI+(THURLBY+THANDAR+INSTRUMENTS)/EL302/displayProduct.jsp?sku=IN01599
which is what I use.

I'm sure you'll be able to make something good, but will it be worth it if an off-the-shelf is better/safer for an extra £20? Please don't think I'm trying to put you off a nice project as I'm lazy.

 

[hippy]

What is the best way to monitor a Home-Made DC 22V Max Variable Power Supply.

For ease of use, simplicity and chapness, I'd also suggest a dedicated panel meter or hacked multi-meter. I picked up a couple of panel mounted current meter displays off eBay whcih were incredibly cheap and just needed a 1K to make work as voltage meters.

http://search.ebay.co.uk/search/search.dll?satitle=panel+meter&category0=&fsop=32&rpr=78

A 30V transformer dropped to use 5V at 1A output will put a lot of heat out so a variable transformer may be better, or a 15-0-15 switched to use as 0-15-30 a second best. You probably won't match a well designed purpose build PSU though, especially if it has good 1mA to 5A current limiting.

One source for 30V 5A transformers and similar are kid's cooking sets. I've found a few secondhand plastic ovens which have a separate plug in PSU which are very useful.

I enjoyed building my own variable PSU so it's a worthwhile project, but if you are simply looking to save money it may be a false economy.

 

[Lliam20789]

So this sort of thing?

http://cgi.ebay.com/3-1-2-Digital-Red-LED-DC-0-20V-Volt-Panel-Meter_W0QQitemZ230183157348QQihZ013QQcategoryZ73158QQcmdZViewItem

and even...

http://cgi.ebay.com/3-Digital-Blue-LED-DC-0-2A-Amp-Panel-Meter_W0QQitemZ140169621965QQihZ004QQcategoryZ73158QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

What exactly did you mean by a Picaxe "controlled over-current trip"?
How would it work, and what sort of sensor would it use?

I think I'll go ahead with the project, I think I'll be able to find most parts so shouldn't be overly expensive...

We've done the basics of power supplies at school so it only the fine tuning and high Amp variable supply I'm not exactly sure on...

Thanks for the input and more suggestions are most welcome!

 

[hippy]

If you're using an LM317 variable regulator, one thing to watch for in design is that if the variable resistor goes open circuit the output jumps to maximum. Especially important if using a rotary switch to select fixed value outputs.

I used 4K7 plus 1K linear pots in series for coarse and fine setting. If you look at output voltage versus R the results are not very linear but a couple of resistors across the pot wipers and ends sorts that out. I'll see if I can find a circuit and what the values are if anyone is interested.

 

[Lliam20789]

I'll see if I can find a circuit and what the values are if anyone is interested.

That would be great Hippy!

I just found A bunch of dead supplies at school that someone has even ordered the replacement parts for, but never got round to fixing them.

There's a new 15 Amp Transformer with 2 * 0V,9V and 12V leads. I also have a LM317k for the variable supply and an LM317 for the 5V. A case from one of the dead power supplies will also be the right size.

Im thinking that a 12V Variable source will be ok.

So all thats left is to put together a circuit and get those Voltage/Amp Panel displays... any input on those?

Thanks again,
Lliam.

 

[Dippy]

Those displays look nice and if you keep below 20V then they look like a direct fit (though I haven't looked at conenction details).
By using a shunt res you can use another one of the displays to show current. You may need an op-amp too if you want to keep the shunt val very low.

Current trip.
You could use an op-amp method to measure the voltage difference across the shunt res. If the current (i.e. Vdrop across shunt) is too high then you could switch off the output.
With PICAXE you could make it quite flashy. Same principle as op-amp but use an LCD display then by adjusting a front panel pot. you could show the preset val of current for trip and the current current (ha).

And don't forget to source a suitable rectumfriar where you will also calculate whether heatsinking is required in 'worst case' mode.


PS. I've just looked at this Data Sheet which gives some nice exampls too:
http://www.st.com/stonline/products/literature/ds/2154.pdf

Note the Application Information and Fig7 gives a nice safe example circuit.
You will have to calaculate your values for your requirements. But also look at Fig 3 which gives some indication that the Vin shouldn't be too high to get best power performance.

You could end up with a pretty clever power supply but it will take some effort.

In the meantime I take it you've got copies of the Data Sheets for the regulators?

As mentioned before, your regulators will have to dissipate a lot of heat so don't forget some 'worst case' calcs on heatsink requirements.

And a bit of suppression on the mains side wouldn't go amiss either.
And fuses all round too.

 

[MartinM57]

rectumfriar ... eh?

Yes, it is found by Google....(Safe for Work)

 

[Lliam20789]

I have two bridge rectifiers, if thats what you mean, with:

MIC
B R 6 2 248

Printed on them, I know they have been used in similar fixed power supplies.

If someone has time it would be helpful to put up a diagram of the hole circuit or the more complex parts; PICAXE etc...
That would e appreciated.

I'm not exactly sure where I should put fuses, I'll have a 3A, maybe 5A on active, but not sure where else.

This may seem silly to you Dippy, but can I just confirm;
The PICAXE circuit would allow you to set a trip current threshold?
In what situation would this be useful?

I know this will take some work (and learning) but I think I'll go as "Flashy" as I can, better to have and not need it yet then to need it later but not have it.

 

[Dippy]

Doesn't seem silly at all.

Here's the scene:
Mk1 PSU. No protection except a 3amp fuse. Lliam is breadboarding his new circuit, but in the excitement a couple of wires are touching.
He switches on the power .... poof!.... a cloud of terminal smoke appears.

Scene 2:
The Mk2 PSU has current limiting. The circuit should only take a few milliamps. Twiddle the knob to 5mA.
Power on. Doesn't work. Hang on.. display says OVERLOAD. Switch off.
Phew, Holy Smoke batman , saved by a current sensing trip.

My TTi PSU has saved many an embarrassing moment by being able to set a max current. It also saves groans of agony and re-ordering.

Up to you. The only thing silly is NOT to have it

"If someone has the time..." - ah if only I did have the time, then I could make one for myself too.

PS. I can't identify that bridge, what package?

 

[hippy]

Current Limiting

Current limiting has 'saved my bacon' a number of times, exactly as Dippy describes; "Danger Will Robinson !!!" flashing in neon is far better than a room full of magic smoke.

It doesn't just help for the case of shorts, but reversed chips, outputs wired to 0V / 5V and anything else which can destroy a PICAXE. It's like a rev-limiter on a car, it simply stops disaster when set appropriately. I keep the limiter at 10mA usually, only going higher if I need to.

One thing a good PSU wil show is what the current limit is set to as well as actual current drawn. Knowing it's gone into current limiting mode is also very useful.

Variable Voltage Output

Apart from ' because you can' it's worth considering what you need a variable voltage supply for an what range. 0V-100V, 25A has a 'wow' to it but is not usually required. Do you really need 30V or 20V even ?

For most PICAXE work, I've needed to go down rather than up, testing at 3V3 etc. Going too far up would destroy the chip.

To avoid 'instant PICAXE death', make sure you have a switch on the output connector side not just mains/input side, and get used to using it. You don't want to turn on and find the knob has been knocked and you've dumped 20V into a PICAXE.

The advantage of your own PSU design is you can add whatever you want. Personally I'd add an option so the PSU would not ever put out more than 5V unless a button were pushed and held for sometime to enable that, and always drop back to a 5V limit on every power up.

 

[Lliam20789]

I see... Thankyou.

So if I have a fixed 5V PS (for PICAXE work) and a variable one, 0V - 12V (for motors, etc.) I should monitor current of both?

I guess I could have a POT and two push button, controlling a little interface with the LCD and the current limiter?

I assume that the current limiter would then simply switch a relay to shut off power?

Oh and sorry Dippy, first time I've seen one so; Its a small black rectangular prism, about 6-mm high, 15-mm wide and deep, with one of the corners tapered. Also has a hole in the center? hope that helps...

Thanks for you help so far!

PS: "ah if only I did have the time, then I could make one for myself too" Bit that way isn't it... I've got exams next week.

 

[Dippy]

In reverse order:-

1. Good luck with exams. Don't let this project get in the way of school/college work ... that's what my Dad would have said.

2. Rectifier. Sounds big and butch. Prism shaped? As in triangular cross-section? Never seen that. Usually rectangular or square with an 'ole in it for clamping/heatsinking. Tapered/chopped corner to I/D orientation/polarity. If it was a square one, with legs at each corner, the taper is usu the + o/p with the -ve being diagonally opposite.
If there are no markings you could work it out with multimeter or a gentle dabble with batteries + LED.
If in doubt buy a new one... complete with Data Sheet.

3. If I were doing this (and I really seriously wish I had the time) I would have separate variable trips on the variable V o/p and the 5V o/p. Measure the Vdrops across shunt Rs, or recently a thread showed a current sensor from Zetec which would make life easier. And yes a couple of Pots to adjust preset and an LCD to display V / Itrips / I. Yes, you could definitely make something really good here.

Switched relays for o/p would be easiest.

One question: would you want your variable o/p to ever go below 1.2V?

Exams first, yeah?

(Fantasy PSU: a nice GLCD which would show all the info and even a power consumption rolling graphic and/or transient detection. Keypad entry for variable V O/P... sorry I'm off in a world of my own now).
I'd really like to design one as a commercial product but there's all the rules and regs/approvals/safety tests so I won't bother. Perhaps if I printed Made in CH*** on the back I could avoid all that.

 

[BeanieBots]

@ Dippy,
Ever wondered why everything comes a wall-wart these days?
Gets round the regs problems. CE (or whatever) marked wall-wart, do (almost) what you like in the box.
If I were doing this:-
60W laptop PSU. 28X using PWMout for current & voltage demand.
Four buttons for I/V inc/dec. AXE033 for demand/actual display.
If feeling brave, hack PSU and fit op-amp+demand+actual into feedback.
If not so brave, home-brew buck-boost switcher. (solves low voltage high current heating problem)

Already started 2nd option for 1A Pb charger / general purpose / solar controller.
Got a few problems with stability when in constant current mode (can't get enough phase angle to overcome output response delay) and haven't touched it for a few weeks but should get time over Xmas to fix.
Might publish when done but not one for the faint-hearted.

 

[Dippy]

BB: "...(can't get enough phase angle to overcome output response delay)"
- perhaps your inertial dampers have gone offline

Personally, for a posh bench PSU I'd avoid s/mode. It's impossible to get dead smooth.
But for a butch general purpose one I would.
Hey a PSU with both.... don't tempt me.

Personally, I wouldn't do pwm controlled voltage unless it was simply to control a pukka linear regulator (as opposed to feedback).
But that's just me being personal.

 

[BeanieBots]

I think you missed my point.
The PWM is to provide DEMAND. (ie, replaces POT in main PSU)
The "PSU" then supplies whatever that demand is.
Why not use both? SMS providing +2-3v into linear.
Headroom voltage on linear used to control SMS demand.
Common enough arrangement for that "smoother" finish.

Thinking about it. You may have inadvertently put me back on track with my instability issue. PWM frequency beating with SMS frequency.
Time to get the 'scope out again.

 

[Dippy]

"I think you missed my point."
- more accurately I misunderstood your point Yes, I understand now.

S/M into linear, still won't be as smooth/clean as linear only. If it is then I'll eat your hat and coat - even if it's a crash helmet and greasy Belstaff

S/M vs linear?
If I wanted a PSU where I was experimenting with, for example, sensitive I/R receivers which required huge gain to pick up a piddling signal s/posed on DC then I wouldn't use s/mode.
However, fiddling around with m/controllers flashing LEDs then a good s/mode is perfect and more efficient.
Horses for Courses.

But if I were to make a single bench PSU for all stuff then it would be linear.
Anyway, I don't really want to dwell as we both know.

 

[BeanieBots]

I agree Dippy.
Anyway, back to the original plot:-

I assume that the current limiter would then simply switch a relay to shut off power?

I don't think that is what was intended. Certainly not what I had in mind.
Rather than switch the power off when the current reaches a certain level, the PSU should switch into constant current mode and only supply that amount of current as a maximum.

Consider the following as a very simple example of what I mean:-
An LM317 regulator can be made into a basic constant current source.
Feed that constant current in to a 7805 type regulator.
When there is no load on the regulator, the current source will go to it's maximum possible voltage (a few volts less than the supply) so the 7805 will have high input volts and put out a regulated 5v.
As the load is increased, the current source will start to operate in its constant current mode. When this happens, the volts droped across it will start to incease and thus less volts are available for the regulator. When the load current equals the constant current, there will be no volts left and the regulator output will drop in volts but it will still supply the amount of current set by the constant current source.
Sounds more complex than it is.
Very simple way to make a regulated 5v supply that can't destroy anything.

 

[Dippy]

Yes, your description is what my TTi PSU does.
A maximum current limit rather than a current trip.
That, I agree, is the best way.

I think someone (hey BB you've got the time ) will have to post a circuit.

PS. I was getting worried I'd have to eat a hat!

 

[BeanieBots]

greasy Belstaff in the post
Is THAT why those OP smoothing devices are called chokes?

There's a rumour that I might have a spare hour this weekend, so you never know!

 

[Tom2000]

60W laptop PSU...

That's a great idea, Beanie. I've got one gathering dust that I've earmarked for a motor control project some day. 19v, 6.3 amps.

That could be the heart of a pretty nice bench supply. For those contemplating a bench supply project, it's something to put on your "scrounge one up" list.

Tom

 

[BeanieBots]

Circuit diagrams galore!
http://cache.national.com/ds/LM/LM117.pdf
No point me re-inventing the PSU.

 

[hippy]

I'll see if I can find a circuit and what the values are if anyone is interested.

That turned out to be easier than expected ... !

http://www.hippy.freeserve.co.uk/psu.htm

The 4K7 and 1K0 pots are the values Maplins happen to sell with nice smooth action at lowish cost. The circuit was tailored to suit those pots and maximum 20V output which is what my analogue meter was marked for.

I did also find some notes ( below ). I recall it's a bit tricky because it's a balancing act between pot values ( Rbot ), Rtop value, Vout and getting a linear pot change to voltage adjustment. I think with 4K7+1K0 it worked out that 20V was optimal anyway and just convenient that the meter matched that.

The fundamental circuit for the LM317T is as shown below ...

             LM317T
           .---------.
        >--| I     O |----.-------- Vout
           |   Adj   |   .|.
           `----.----'   | | Rtop
                |        |_|
                |         |
                `---------|
                         .|.
                         | | Rbot
                         |_|
                          |
        >-----------------^-------- 0V

The maximum output 'Vout' can be calculated using the following equation ...

                     /    Rbot \
    Vout  =  1.25 x ( 1 + ----  )
                     \    Rtop /

Likewise, the value of 'Rbot' for maximum output voltage can be calculated as below ...

              / Vout    \
    Rbot  =  (  ---- - 1 ) * Rtop
              \ 1.25    /

As recommended by the datasheet for the LM317T, Rtop should be 240R or less.,

 

[Lliam20789]

Wow, thanks for all your help!
I'm sorry I do not have time to give feed back to all you replies, but basically this is a great help!

One major aspect I am unsure of is how the current regulation part works? As I am trying to keep cost low and design not overly complex, I do not see exactly how this works.

Would you agree that I am safe to start, (IN MY SPARE TIME...) a basic rectifier, filter circuit?

Thanks again!