Electrical/Electronics wiring circuit HELP! Need to convert 0-150Vdc 10 ~15vdc

[Gillbot]

I have a variable voltage DC buss (0v to ~150V) and I need to pull 12-18v off of it @ ~5-10 amps. Any ideas?

It's actually a car alternator sans the regulator circuit. I'd like to tap off this unregulated section after the rectifier to supply the alternator field with manually variable DC from 0~18vdc to vary the current output of the alternator itself. The down side is, I need to keep the output of the alternator unregulated voltage wise.

 

[KalTorak]

Why? What's the bigger picture of the thing you're trying to make?

 

[Gillbot]

Simply a welder. I't been done many times over but people usually use a battery to power the field then have to recharge it over and over or resort to an AC adapter. I want to make mine 100% cord free and portable w/o having to worry about recharging a battery to use it.

EDIT> or let this unit recharge the battery itself.

 

[Qacer]

Are you going to create your own voltage regulator circuit?
The simplest way to tap off 18v or so from your 150v DC bus is using a voltage divider, but that itself isn't very reliable due to voltage drops across the internal / line resistance at around 5-10A. I've only had experience with linear power supplies. You can re-create a simple variable supply using schematics from Radio Shack's power supply handbook. If you do, then you'll probably need a power transistor with a high power disappation coupled with a heatsink. The idea is you would use the DC bus as your main power source and your custom power supply would sample and correct the output voltage to your 18v spec. The output current would depend on your load, so at 18v, your load should be around 1.8 ohms to get 10A.

 

[KalTorak]

Something screams "perpetual motion machine" here, it seems. The alternator's rectified output creates the electric field that the alternator needs to provide a rectified output??

Anyway, if I black-box that out of the way, the rest is how to step an unregulated high-voltage DC input down to 15VDC out. It's not clear to me if the high-voltage input is always known to provide >15V - if it is, then this isn't a hard problem. (Use a buck-topology switching DC-DC voltage regulator.)

I've never done this for the case where the input could be less than or greater than 15V. I'm sure it's doable, but never had cause to look into it.

 

[Mark R]

Linear based regulators are probably out of the question given the voltage and current levels required. Switching regulators are difficult to design and build - and a trivial mistake can incinerate lots of components - I wouldn't recommend anyone other than an engineer with experience in switching PSU design try to build a high-voltage regulator.

The simplest way I can see is to try and use an old PC PSU.

You just need to connect the PSU input direct to the AC output of the alternator. Unfortunately, most PC PSUs need an input voltage of about 310-350V DC and this comes from a voltage-doubling rectifier - so they won't run directly off a 150V DC supply. Many alternators are 3 phase, so you may need to change the rectifier from a single phase to a 3 phase one.

If you are feeling really brave, and have a suitable knowledge of electronics - it shouldn't be too hard to modify the feedback control loop to allow a potentiometer variable output voltage. Some reverse engineering of the feedback resistor networks will be necessary.

Just another thought - you may find that a PSU with 'active PFC' could run on unregulated 150V DC - the 'active PFC' circuit is actually a complete PSU in itself - it takes unregulated DC (actually unsmoothed rectified AC 0-340V) and steps it up to about 380 V regulated DC. (That's why active PFC PSUs often have lower efficiency than more basic PSUs). You shouldn't even need any internal modifications - I'd expect that you could just hook the 150V DC up to the input.

Note, that it any case I don't know how well the PSU would cope with big sags in the input - as might happen if the alternator is heavily loaded. Designing a PSU to cope with that is actually very difficult. Although, if you have a battery for the field coil - then this should be able to keep the alternator running if the PSU drops out. Note though, that ome PSUs may trip-off and need power-cycling if they shut down due to a 'brownout'.

Note, that I've not tried any of the above techniques - so try them at your own risk.

Could you not use 2 alternators? One to provide the field current for the other?

 

[Geniere]

Hmm- the last time I took an alternator apart, I found it was 3 phase, but the rectifiers were built in. In auto use, a typical alternator can provide about 14vdc at 100 amps DC. A PC power supply cannot be used on the output of the alternator nor can I figure out why one would want to, as the alternator output is already DC. It could be used to supply the field voltage to the alternator and swing the field voltage though a range of 24vdc, but I?m unsure if the negative 12vdc could support the current draw. The input voltage to an AXT PSU in the US is 120vac, not 310-350vdc.

Welding voltage (MIG) would be about 30vdc although AC can be used for some applications. The normal current for welding would range from about 25amps to over 250 amps depending mostly on the thickness of the metal to be welded. I?ve never investigated Arc (stick) welders, so I can?t comment on that. At a high price of about $500.00 there are MIG welders available to attach to ones auto battery.

Can a single alternator be made to work? Probably yes, but it would likely restrict welding to thin sheet metal. If I were to do it, I would use a bank of three batteries and switch between 24 and 36vdc operation.

With 90 amp MIG or Flux welders available in the $125 dollar range (Chinese versions from Harbor Freight), I would not think it worth the effort.

For the record, I?ve welded exactly once in my entire life.

 

[Gillbot]

Don't get hung up on the fact that it is (was) an alternator since much of it's guts were removed long ago. I have removed all the rectifying and regulating pieces of the alternator so it produces 3 phase AC out. I have an external 3 phase rectifier (from a toshiba VFD drive) converting the 3 phase AC to DC.

What I want to do is tap off of the rectifier's output to feed the alternator field. This will allow me to manually adjust the field voltage from 0 to ~18v DC to regulate the output current of the alternator itself. This will in turn produce a varied output voltage at the rectifier so it needs to be regulated and fairly high amps since the field will pull a few, although I have not measured it.

I know I can easily use an external source but I want to make this thing completely portable and self sufficient meaning no AC adapters. Batteries are ok IF I can get this output so that the batteries will be recharged automatically.

 

[Geniere]

I know I can easily use an external source but I want to make this thing completely portable and self sufficient meaning no AC adapters. Batteries are ok IF I can get this output so that the batteries will be recharged automatically.

I?m not sure what you mean by portable. How do you plan to drive the gutted alternator? If by electric motor, than you would need a battery(s). If you have to drag batteries around, than why not eliminate the alternator. If you want to use a small gas engine or your vehicle, that?s a whole ?nother story. By what means would you charge the batteries? If you can be more specific, I or others, may be more helpful.

 

[Gillbot]

It's driven by a Briggs gasoline engine.

Let me get a picture and you'll see the basics.
http://cofba.org/users/gillbot/Welder/smP1010319.JPG

EDIT> The picture you see is before damage. This is the new rectifier section. BTW, those caps are not connected to anything, they are just there for visual impact.
http://cofba.org/users/gillbot/Welder/P1010329.JPG

This is meant to be a 100% portable unit and it a project to see how cheap I can do this. I was going to put it into my 4x4 but I decided not to as I'd rather have it available anywhere. It started life an an old mower and a 100A GM alternator but upon removing the case, I destroyed the rectifier. I got one from a toshiba VFD at work so I made it externally rectified. I also removed the regulator so the voltage will be high enough to weld with. Most people who do this just use a motorcycle or car battery to run the field, I'd prefer to make it self sufficient so I don't need to recharge it or anything, just gas and go. Later on IDEALLY, I'd like to replace the regulator and make a switching circuit so that it can be used as a welder in one mode and a gas powered battery charger/jump starter on the other. I can also add an AC power box to make it a portable generator also.

This is just the test/trial run to see how well everything goes and works. Later on, I plan to upgrade to a larger electric start motor and much higher output alternator. I am also considering electric clutches so that it could be used to drive either the welder/generator (alternator) or an air compressor, depending on which clutch is engaged.

 

[Geniere]

To get full wave rectification, you need 6 diodes, with (ballpark) 150vdc, 200 amp rating.

It should be possible to use the remnant magnetic field of the stator, and feed back voltage from the output, to excite the field winding. The circuitry for that would take a great deal of thought, so I?d rule that concept out and use a battery for field current.

The easy way to do it would be to ignore the need to charge the battery in the field. I would guess the field current would be about 7-8amps, so a full charged auto battery should last quite a while. There is a way to charge it in the field, but not weld and charge at the same time, at least not easily.

I?d guess the mower engine gets max torque at around 3000rpm, so that?s where I would run it with a 1:1 mechanical drive to the alternator. If the alternator has an internal regulator remove or bypass it and connect the battery terminals to the field windings through a 10amp fuse. We?ll worry about regulation later. Configured this way, I think the DC output would rise to 75-100 volts (guessing). That should get you welding albeit with little control. It might be sufficient to vary the output by adjusting engine speed. If not?

Then you need to insert a variable voltage regulator between the battery and field winding fuse. The regulator will have to be able to handle the field winding current. You can run the engine at max torque rpm and use the regulator to adjust out put. You can also re-charge the battery by connecting the alternator output to the battery and aadjust for 14vdc.

 

[Gillbot]

Whay are you so concerned with the regulator and the rest of the system? I have all that stuff handled via a THREE PHASE rectifier from an industrial drive which has SIX diodes. I also have a 6" pulley on the motor and a 2" pulley on the alternator. This will achieve ~6000 rpm to provide max output from the alternator. There also is NOT enough residual magnetism in this to get the circuit going so is will need a small battery to get it going, then the tap off of the main line will provide the full field power.

I didn't ask for help with designing this thing, I need help with the power circuit. NO, I will not resort to a battery for a full time field power supply.

 

[Mark R]

I think you're going to struggle powering the field coils from the main DC output.

A simpler solution might be to utilise the fact the alternator is 3 phase. Use one of the phases to support a battery in the supply of field current. The other 2 phases can be rectified and used to provide the arc output.

You will still need a suitable regulator on the field supply - but construction will be easier because the alternator output will be a bit more consistant. Options are a difficult to design and construct switching power supply (technically the best solution), or a linear regulator (much easier to build, but lower efficiency and lots and lots of heat - massive heatsink-fan a requirement). I suspect that a linear regulator will also need a transformer to step the voltage down - I cannot tell you what specification transformer you would need - this would have to be determined empirically.

 

[Geniere]

Good luck!

 

[davesaudio]

I've forgotton most of my motor/gen theory but the hacker in me wonders
are we looking at this the right way?
how about an adjustable current regulator to feed to field?
tried a suitably sized power resistor alone?

 

[Gillbot]

Originally posted by: davesaudio
I've forgotton most of my motor/gen theory but the hacker in me wonders
are we looking at this the right way?
how about an adjustable current regulator to feed to field?
tried a suitably sized power resistor alone?

That could work but the problem is the voltage will vary enormously between open circuit and welding. As soon as you strika an arc the voltage will sag, this in turn would drop the current through the resistor and into the field, thus dropping the current more, etc, etc.

If you can maintain a constant voltage, the field current will remain constant independent of what the main DC line is doing.

 

[davesaudio]

so a resistor is NG
what matters here - current reg or voltage reg?
muses to oneself in public:
I cant think of any 3 terminal 5A regs that could handle such a high voltage input.
that would be too simple
heat dissipation would be an issue anyways
so back to pwm switching
have u poked arout somewhere like marlin jones
http://www.mpja.com/
to see if they've anything suitable?

 

[Gillbot]

DK1501-7R or CK1501-7R from http://www.power-one.com will be my best bet I think.

 

[NeoPTLD]

The alternator regulator doesn't vary voltage to field coil. It monitors output voltage, then gives a constant voltage variable duty cycle(also known as PWM) current to filed coil at i think somewhere around 400Hz.

I'm not sure how it's all hooked up, but you should be able to change the output voltage by tampering with the feedback input to regulator.

For example, if you use a resistive divider into the input of feedback, the voltage going into regulator will be lower than actual voltage.

If the resistive divider divdes it by 1.2 you can fool it to put out 16.56V. Because the input to regulator is 1/1.2 of the actual voltage it will "think" there's 13.8V(the preset voltage) at output. You can use a variable resistor to change the dividing coefficient.

 

[Gillbot]

Originally posted by: NeoPTLD
The alternator regulator doesn't vary voltage to field coil. It monitors output voltage, then gives a constant voltage variable duty cycle(also known as PWM) current to filed coil at i think somewhere around 400Hz.

I'm not sure how it's all hooked up, but you should be able to change the output voltage by tampering with the feedback input to regulator.

For example, if you use a resistive divider into the input of feedback, the voltage going into regulator will be lower than actual voltage.

If the resistive divider divdes it by 1.2 you can fool it to put out 16.56V. Because the input to regulator is 1/1.2 of the actual voltage it will "think" there's 13.8V(the preset voltage) at output. You can use a variable resistor to change the dividing coefficient.

As I said before, don't get hung up on the fact that it WAS an alternator. Any remnants of the electronics in there are LONG gone. Not it runs 100% unregulated ALL the time full output. The current output will be regulated via controlling the voltage on the rotor.