Help with Schematic Analysis?

[TBennettcc]

I received an old adjustable DC power supply. (The plate on the top reads: D.C. Test P.S. / MOD PS-1 / 0-50 VDC 4 AMPS / 115 VAC PRI 60 HZ)

I didn't feel it was working correctly, so I tore it apart and made a schematic. I need some help in deciphering what exactly this was designed to do, and how it is supposed to work. (I have no idea if this wiring is factory, or if the wiring is "correct"; I have a feeling someone has messed with this before, as some of the wires and switch contacts are in need of replacement. I don't know if any circuits have been modified from what they might have been.)

According to the attached schematic, I'm good with most of the beginning:

AC 120V in, power indicator lamp (LMP1), then the fuse (F1) (I'll switch those two around when I rebuild this... not much sense having an indicator light for line voltage even if the fuse is blown...) The next part (VAR1) is a Type 10B Powerstat Variable Transformer (AKA, Variac), rated 2.25 amps, 297 VA. (I didn't have the right schematic part in EAGLE CAD, so I made do with a center-tapped transformer. You get the idea.) Next is a two-to-one transformer (TR1). Next we have the bridge rectifier (BR1).

This is the part where things get fuzzy. The rest of the schematic is a bunch of resistors connected to switches. I talked to a friend, and he said that the resistors connected to SW2 between poles 0 and 1 are to limit current. Okay, fine, but how does that work if my AC (and thus my DC) is finely adjustable? I could see that maybe with a single input voltage, but how does this work over the range?

Also, R3 is simply a piece of wire loosely coiled up (3 turns, .75" in diameter by 1.5" in length when coiled). I can't imagine what purpose this could have served? When I opened up the unit, there was a char mark on the metal plate directly above this wire. It's possible I could have caused this by plugging the unit in and turning knobs without knowing what those two rotary switches were connected to at the time.

Resistors R4-R7 are 10ohm, 10%, 2 watt wirewound resistors. With both contact points not connected at the rotary switch, I measured the resistance across R4, and get a value of 718 ohms??? Even if Ohm's Law is followed, those resistors are in parallel, which means the resistance should go down (to 2.5 ohms, if my math is correct), not up, and certainly not to 718 ohms!

The plate on the unit states a maximum ampacity of 4 amps, but I was easily able to exceed that number using an ammeter connected across the final VDC + and - terminals and just turning the Variac up. Granted, that piece of wire (R3) almost caught fire, so I quickly turned it down.

The rotary switches also seem to play a role in adjusting the meters on the front. Both meters appear to be the same. The scale on both meters goes from 0-100. There are no other markings on either meter, aside from a stamp on the back of each molded into the plastic from the Simpson Electric Company. The needles move on the meters, but not accurately. Any ideas for these?

This would be an awesome troubleshooting tool for me if I can get it to work. How do I make this work, and safe (i.e., no risk of electrical fires)?

I apologize for the book. I wanted to be clear. Any and all help is greatly appreciated.

Thank you for your time.

 

[(*steve*)]

The switch positions seem to do no more than change the current and voltage ranges of the meters. There is no regulation of voltage or current.

 

[poor mystic]

looks like a battery charger

 

[Resqueline]

What Steve said. It's only an adjustable voltage power supply, without any kind of overcurrent protection. It could use a fuse on the output.
R3 is a resistor, probably less than 1 ohm. Car heater fan speed resistors are often made that way. It sounds like R4-7 are fried.
It can be fixed and used as-is, but the meter scales will be cumbersome to use & read, and the A-meter & shunt resistors are prone to damage.
The alternative is to build a fully regulated electronic PSU out of it, with voltage & current controls. One example is being discussed on the board here right now.
The variac can then be used as a manual "headroom" control for the electronic regulator, to limit/reduce power dissipation when drawing high currents at low voltages.

 

[TBennettcc]

What Steve said. It's only an adjustable voltage power supply, without any kind of overcurrent protection. It could use a fuse on the output.
R3 is a resistor, probably less than 1 ohm. Car heater fan speed resistors are often made that way. It sounds like R4-7 are fried.
It can be fixed and used as-is, but the meter scales will be cumbersome to use & read, and the A-meter & shunt resistors are prone to damage.
The alternative is to build a fully regulated electronic PSU out of it, with voltage & current controls. One example is being discussed on the board here right now.
The variac can then be used as a manual "headroom" control for the electronic regulator, to limit/reduce power dissipation when drawing high currents at low voltages.

Care to point me to the post you are referring to?

Also, looking at poles 0 and 1 of SW2, aren't those (sets of) resistors being inserted in series with the output? Wouldn't that be some form of current limitation?

So, the general sense of what I'm getting from everyone is that without further modification, this thing is dangerous to use?

With something like this, is there any way to tell what the theoretical current limit would be, with no resistance in the circuit (i.e., let's assume a dead short across the final DC+ and DC- terminals,) taking into account that the Variac says its current limit is 2.25 amps max? I got a cheap digital multimeter "capable" of measuring up to 10 amps, set it to measure amps, put it across the final DC + and - terminals, and turned the power supply up fairly quickly, and then down just as fast when I noticed the "1 ohm" resistor start to glow (and my probes got pretty warm, as well). I think I got it to 8 amps. But I guess that's what you're talking about, that if it was designed properly (i.e., proper current limitation), I shouldn't be able to do that, because if I did, and I left it like that, it would have definitely started an electrical fire... So that's what circuit breakers are designed to do. You have to have a load (i.e., a resistance of some kind) between electrical contacts... otherwise, it's a dead short, and you would have an infinite amount of current flowing, and things would heat up quick, fast, and in a hurry. Things that make you go hmmmmm...

To keep the highest current at 4 amps at the highest voltage of 76 VDC, I would need a 19 ohm resistor (to make things simple, choose an E12 value of 18 ohms, which would result in the max voltage of 72 VDC.) Maximum power possibly dissipated across this resistor would be (72*72 / 18 = ) 288 watts... yikes. No wonder this thing caught fire...

Well... I really don't need 76 volts... if I limit the power supply to 25 volts, the resistor would need to drop 51 volts (76 - 25), so we'll choose a 13 ohm resistor... which will need to dissipate 200 watts... :-/

...where was that nice regulated electronic power supply again? >.>

I've actually been looking into some tutorials on designing switched-mode power supplies. Any suggestions?

While I was in college, I did a project using pulse-width modulation. I had a power supply feeding the emitter of a transistor, the collector hooked to ground, and the PWM signal connected to the base. I was controlling a small DC fan motor.

When the fan motor was hooked directly to the power supply, the blades were hard to stop (the motor had good torque). However, when I was controlling the fan using the PWM circuit, the fan would stop spinning much easier (much less torque). Can anyone explain why?

I have received a lot of knowledge and understanding about electricity and electronics in the past few days, being here on this forum. Everyone has been amazing. Thank you so much. I hope to keep learning just as much in the future.

Thank you for your time.

 

[(*steve*)]

Care to point me to the post you are referring to?

I believe it's my post about four posts above this one.

Also, looking at poles 0 and 1 of SW2, aren't those (sets of) resistors being inserted in series with the output? Wouldn't that be some form of current limitation?

They are current limits to pretty much the same extent that pedestrians on the road are speed limits. Sure they provide some sort of theoretical limit, but you're going to kill lots of stuff if you rely on it.

They appear to me to be performing the function of shunts across the meter to give various current ranges.

So, the general sense of what I'm getting from everyone is that without further modification, this thing is dangerous to use?

Not so much dangerous as not a general purpose tool. I'm sure it was quite adequate for whatever it was designed for. However what it is NOT is a regulated general purpose power supply.

With something like this, is there any way to tell what the theoretical current limit would be, with no resistance in the circuit (i.e., let's assume a dead short across the final DC+ and DC- terminals,) taking into account that the Variac says its current limit is 2.25 amps max?

Huge. And almost certainly enough to fry stuff including the resistors, the switches and the rectifiers. If you didn't stop there, the transformer, the variac, and connecting wires... As I've said. No current limit.

Practically, the current would be limited to some value, 10A to 20A maybe.

I got a cheap digital multimeter "capable" of measuring up to 10 amps, set it to measure amps, put it across the final DC + and - terminals, and turned the power supply up fairly quickly, and then down just as fast when I noticed the "1 ohm" resistor start to glow (and my probes got pretty warm, as well).

At that point you've got a few indications that bad stuff is happening. And I really don't think you need me to tell you that.

I think I got it to 8 amps. But I guess that's what you're talking about, that if it was designed properly (i.e., proper current limitation), I shouldn't be able to do that, because if I did, and I left it like that, it would have definitely started an electrical fire...

A properly designed regulator might have shut down, or gone into current limiting, or done *something* that wouldn't result in its own demise, and preferably would have saved your equipment too.

To keep the highest current at 4 amps at the highest voltage of 76 VDC, I would need a 19 ohm resistor (to make things simple, choose an E12 value of 18 ohms, which would result in the max voltage of 72 VDC.) Maximum power possibly dissipated across this resistor would be (72*72 / 18 = ) 288 watts... yikes. No wonder this thing caught fire...

heh, no wonder

Well... I really don't need 76 volts... if I limit the power supply to 25 volts, the resistor would need to drop 51 volts (76 - 25), so we'll choose a 13 ohm resistor... which will need to dissipate 200 watts... :-/

And here is the problem with a linear power supply. The excess voltage (at the max current) gets turned into heat. So you really want to keep that excess voltage to a minimum.

...where was that nice regulated electronic power supply again?

I've actually been looking into some tutorials on designing switched-mode power supplies. Any suggestions?

Switchmode regulators are non-trivial and I would not recommend them as your first power supply project.

Let me give you another car analogy. It's a bit like saying "Hey I'm having trouble learning to steer and change gears at the same time in my car. I hear that Formula 1 cars have the gear change on the steering wheel. Would it be easier to start my learning to drive one of these?"

A better place to start would be a simple fixed voltage power supply. Maybe something that can deliver 5V at 1A. Then based on that (even just the design exercise of that) work up to something a little more ambitious, say a 2V to 20V 1A regulated power supply.

Once you start talking about variable power supplies with either an extended voltage range, or high current then you're also talking serious issues to be considered. Mistakes can be more expensive, and occasionally pyrotechnic.

While I was in college, I did a project using pulse-width modulation. I had a power supply feeding the emitter of a transistor, the collector hooked to ground, and the PWM signal connected to the base. I was controlling a small DC fan motor.

When the fan motor was hooked directly to the power supply, the blades were hard to stop (the motor had good torque). However, when I was controlling the fan using the PWM circuit, the fan would stop spinning much easier (much less torque). Can anyone explain why?

Essentially PWM involves turning the power on and off very quickly. The average power to the device is thus determined by how long the device is on compared with how long it is off. (essentially switchmode power supplies work in a similar way).

When the duty cycle is low (say 10%) the motor is only pushing for 10% of the time and coasting the remaining 90%. Thus, it is easier to stop.

That's not the greatest explanation in the world, but it should convey the general idea.

I have received a lot of knowledge and understanding about electricity and electronics in the past few days, being here on this forum. Everyone has been amazing. Thank you so much. I hope to keep learning just as much in the future.

Thank you for your time.

Always a pleasure

 

[TBennettcc]

...The alternative is to build a fully regulated electronic PSU out of it, with voltage & current controls. One example is being discussed on the board here right now...

That's actually the part of Resqueline's post I was referring to. Sorry if that was unclear.

Thanks for the reply, Steve.

...
A better place to start would be a simple fixed voltage power supply. Maybe something that can deliver 5V at 1A. Then based on that (even just the design exercise of that) work up to something a little more ambitious, say a 2V to 20V 1A regulated power supply.
...

I'd say 20 volts at two amps max would probably be the most I would ever need, at least for now, considering I'm mostly working with and trying to repair small stuff. So what you're talking about is perfect. Have any suggestions as to where I might start, or where I might find a project like that that is easy on beginners?

Which brings me to another project. I'm looking for a circuit that can handle probably an amp of power at 12 volts (input), and will give me as much current as possible (maybe 1.5 A?) at 7.2 VDC (I need at least 8.5 watts), and can fit in a small enclosed space (1.5" x 4" x 0.5") with no ventilation. Can what you're talking about do that?

...
Essentially PWM involves turning the power on and off very quickly. The average power to the device is thus determined by how long the device is on compared with how long it is off. (essentially switchmode power supplies work in a similar way).

When the duty cycle is low (say 10%) the motor is only pushing for 10% of the time and coasting the remaining 90%. Thus, it is easier to stop.

That's not the greatest explanation in the world, but it should convey the general idea.
...

When I noticed that, I could swear I remember the duty cycle being high... like 90-100%. I thought it was odd, and maybe there wasn't as much "power" (or something) transferred to the motor while it was being controlled by PWM, rather than being directly controlled by an adjustable DC voltage. Oh well. That was over a year ago, and I don't remember much about the project anymore. But I'm sure I'll be learning more about it.

As always, Steve, thank you very much for your time.

 

[davenn]

ok here's a pretty straightforward variable PSU 1.2 - 30V variable and up to 1.5 Amp

to increase current capability up to 2 to 3 Amps you could subsitute the LM317 for an LM338 which can handle up to 5 Amps when well heatsunk.
If you did that then change C1 from 1000uF to ~ 4700uF to 10000uF and change C3 to 10uF. And ensure that the transformer is rated for at least 5Amps.

cheersd
Dave

 

[TBennettcc]

Thank you very much, Dave.

I found the website you got that from, looks very interesting! And, he's a ham radio operator to boot! (My ham radio call sign is KJ4MUY. Probably be changing it soon; depends on how much of the material I can absorb and understand to pass my Extra exam. I'm a General class right now.)

For future reference, the main website can be found here:

http://www.sentex.ca/~mec1995/

And the Variable Power Supply can be found here:

http://www.sentex.ca/~mec1995/circ/VarReg1/VarReg1.html

 

[TBennettcc]

I would like to thank everyone for their help so far on this project. This has been quite an education, and of course, I have more questions!

It really helped when I finally looked up just exactly what a shunt resistor for an ammeter was. Then it was like "...OH! I GET IT!"

So, just to make sure I get it, I'm going to set up a scenario. Tell me if you agree.

The shunt resistors are used as a current divider in parallel with the ammeter. Nothing more. Different resistor values allow the meter to be used over a wider range of current values.

I have attached a 'sub-schematic' of my power supply. This was made in LTSPICE (interesting program, by the way). This sub-section would be as if the triple-pole selector switch was turned all the way to the bottom.

With the resistors totaled up, the total resistance for the circuit is about 106 ohms (the coil of the D'arsonval ammeter after R9 has a resistance of about 10 ohms). At a potential of 5 volts, I calculate the current flowing through this sub-circuit to be about 47 mA. I have connected a second ammeter across the final terminals of the power supply and have measured about 47 mA.

My question is:

When I have a voltmeter also connected across the final terminals, and I connect the ammeter, why does the voltmeter drop (almost?) to 0? Something tells me that I have brought the two sides of the circuit to (near) the same potential, but if that's the case, then why do I still have (47 mA of) current flowing, as this would still take 5 volts of potential to achieve, taking into account the resistance of the circuit has not changed?

Also, re-evaluating the power supply, the only thing dangerous that I can see (taking into account proper use of the device) would be if a short circuit happened when the voltage is cranked all the way up, AND the three-pole selector switch is set to the 1-ohm coil of wire. If too much current is drawn, this sucker will catch fire. Therefore, I propose to put a 5 amp resettable circuit breaker in series between that coil and the meter. Then, no more fires.

How does that sound?

Also, why is a regulated power supply so important? Why would a regulated power supply be recommended over something like this (which, from what I can tell, isn't exactly a linear power supply, and so isn't wasting that much energy in heat?)

As always, thank you for your time.

 

[davenn]

My question is:

When I have a voltmeter also connected across the final terminals, and I connect the ammeter, why does the voltmeter drop (almost?) to 0? Something tells me that I have brought the two sides of the circuit to (near) the same potential, but if that's the case, then why do I still have (47 mA of) current flowing, as this would still take 5 volts of potential to achieve, taking into account the resistance of the circuit has not changed?

Hey Tim

my resopnse Q would be how did you connect the Ammeter?
Its sounds like you connected it directly across the output... ie in parallel with the Voltmeter ?
If so, you have basically shortcircuited the output of the PSU, hence why the Volts dropped to 0. You still got a small mA reading, probably cuz of the difference in voltage drop across the Ammeter compared to that across the meter's internal resistance. A Voltmeter goes across the output of the supply, ie. it is parallel to a load, an Ammeter is in series with the load. You had no other load other than the Voltmeter.

cheers
Dave

 

[TBennettcc]

...but if the ammeter is the only load, wouldn't it be considered in parallel AND in series at the same time? Or do I need a resistive load in addition to using the ammeter to make an accurate measurement?

 

[poor mystic]

Hi Tim
The ammeter can only provide a meaningful measurement when a load is connected.