DC Power Polarity Auto-detect

[Groovenut]

Hey everyone!

So this caught my eye and I did a bit of research but didn't find much in the way of how this might be achieved. The unit that sparked my interest is the Taurus Amps Tux pedal. On the pedal and in the manual it states 9-12VDC Polarity Auto-detect enabling both center negative and center positive supplies. It is also using an internal DC to DC converter to double the input DC voltage. Providing there's a somewhat simple solution, this would take one of the most common repairs out of the inventory.

Anyone have any ideas?

Using a relay comes to mind, but I'm hoping for other solutions.

TIA

[marcelomd]

A simple rectifier bridge? The downside is two diode voltage losses.

[ElectricDruid]

I think that's why they've then gone with the DC-DC convertor. The only downside to a bridge rectifier is the double drop. If you then boost the voltage, that stops mattering. Then you've got something that doesn't care about polarity and still has plenty of headroom. The downside of that is just the complexity.

I do wish no-one had ever run a pedal off a 9V battery. If we'd started with a decent +/-15V supply way back in the day we'd never have got in this mess, and we'd have better S/N and more headroom, as well as much more compatibility with synths and studio gear.
There were early pedals that went this way - the first Roland Chorus, I think, because it was pulled from an amp design and stuck in a box of its own.

Oh well, too late now.

T.

[lietuvis]

It could be just bridge rectifier. I have wired them at work for ac or dc input for dc/dc step down converter using to power cctv cameras, that way I can power them either from ac or dc on input and as I remember there was no difference in polarity when connected from dc. Your  dc/dc converter has to be with isolated output.
That would be my gues

[Groovenut]

Hmmm so something like this?



The switches are just an easy way for me to test in the sim, otherwise it likes to throw errors. 23mA current load.

So a bridge built with Schottky diodes would save a bit of voltage but like was stated earlier, that's not really an issue with the DC to DC upverter.

There's probably a way to do this with Mosfets and avoid the diode drop. Time to experiment.

[R.G.]

I went through this line of reasoning back in ... 2000? 2001? I'll have to look it up.

I was working on polarity protection with a MOSFET, later a bipolar, and headed off onto "make the polarity not matter".  I thought of using a full wave bridge, but realized that while this will work fine for a single pedal with a single power supply, it won't work for more than one pedal sharing a power supply, at least not in all cases.

That's because full wave rectification introduces two diode drop offsets, one on the + side and one on the - side. So "ground" on the pedal(s) carries a DC offset from any pedal on the same power that doesn't have a full wave rectifier. That will cause at least pops and maybe worse.

But technology has moved on. Today there are schemes and even ICs to drive four MOSFETs in an active full wave rectifier bridge. this can possibly reduce the ground offset to tens of millivolts. On the down side, you have to pay for four MOSFETs; this was a problem back when they were $0.50 to $1.00 each. Today, it might be doable.

You can still use a polarity sense and click a relay. That runs the offset down into sub-millivolts, which is probably not going to cause a problem. The relay will cost $2 to $3.50, though, and you need a GOOD polarity sensor circuit.

Then there's the "drop back and punt" strategy. In this form, you put a full wave bridge on the incoming power supply jack, run that into an isolating power supply converter. You get a self isolating power supply in the pedal, so polarity and grounding issues vanish. But you have to pay for or design and do a good job of building a power converter capable of >quietly< running the pedal.

[Groovenut]

Thanks RG!

Here's my relay based idea



With "standard" polarity the diode will block current and the relay will remain deactivated. With "reversed" polarity current flows through the relay coil, engages the relay and swaps the + and - circuit power rails.

Thoughts?

[EBK]

I've said it before, but I still think the best polarity/over-voltage protection is deterrence.   

[ElectricDruid]

I've said it before, but I still think the best polarity/over-voltage protection is deterrence.   

So include a small spark-activated Molotov %^&*tail so that plugging in a pedal with reverse polarity is only a mistake anyone ever wants to make once? Am I on the right wavelength?

[EBK]

I've said it before, but I still think the best polarity/over-voltage protection is deterrence.   

So include a small spark-activated Molotov %^&*tail so that plugging in a pedal with reverse polarity is only a mistake anyone ever wants to make once? Am I on the right wavelength?

It's spelled "cocktail", but you got the general idea.   

[PRR]

> With "reversed" polarity current flows through the relay coil, engages the relay...

..."eventually". Several milliSeconds.

Many pedals will stand reverse for a while, even forever (though not working right).

But if you have something which blows with an instant BANG, relays are slow.

Yes, a second level is another relay to keep power off the circuit for many milliseconds, or until polarity is correct.

[ElectricDruid]

I've said it before, but I still think the best polarity/over-voltage protection is deterrence.   

So include a small spark-activated Molotov %^&*tail so that plugging in a pedal with reverse polarity is only a mistake anyone ever wants to make once? Am I on the right wavelength?

It's spelled "cocktail", but you got the general idea.   

I had to quote your reply to be able to see how you'd defeated the filter and I hadn't. Cunning.

[bluebunny]

I had to quote your reply to be able to see how you'd defeated the filter and I hadn't. Cunning.

It's a trick well-known to the residents of Scunthorpe.

[EBK]

I had to quote your reply to be able to see how you'd defeated the filter and I hadn't. Cunning.

It's a trick well-known to the residents of Scunthorpe.

I learned it here:
https://www.diystompboxes.com/smfforum/index.php?topic=118278.msg1114087#msg1114087
It's quite a useful trick.

[Groovenut]

> With "reversed" polarity current flows through the relay coil, engages the relay...

..."eventually". Several milliSeconds.

Many pedals will stand reverse for a while, even forever (though not working right).

But if you have something which blows with an instant BANG, relays are slow.

Yes, a second level is another relay to keep power off the circuit for many milliseconds, or until polarity is correct.

Possibly still use a series schottky diode after the relay to solve this issue?

[Groovenut]

I looked into RGs suggestion a bit more and found this little goodie



A quick search for the LT4320 and perusal of the datasheet looks like it should be possible (though not the most efficient) to use the lowly 2N7000 for the N channel mosfets. Chip cost is $8 in singles. The LT4320 seems pretty new and Analog Devices appear to be front runners in this tech. Hopefully as these chips are successful more manufacturers will come on board and the price will drop.

From my understanding of the datasheet, this would allow (with the use of a downstream voltage regulator) any voltage or polarity AC or DC power supply (within reason of course) to be connected and still deliver an "acceptable" DC power to the circuit. This could eliminate most power supply related repairs.

I may have misinterpreted the data sheet, so if someone sees something I missed please correct me.

Always good to have options eh

[reddesert]

Speaking of deterrence: I think the series protection diode that prevents a circuit from working if you plug in the wrong power supply is pretty good. It forces the user to try to get the right power supply (I suppose they might send the pedal back because "it doesn't work," but at least it's not fried). The problem with just one effect that accepts an incorrect power supply is that it lulls the user into complacency. Then they try to use the supply with another pedal that lacks this capability, or lacks power protection, and at best "it doesn't work" and at worst they let the magic smoke out.

Of course, accepting a wider variety of supplies is technically superior, but in music equipment, no equipment is stand-alone.

[PRR]

> use the lowly 2N7000

ON-resistance is like 5 Ohms. Much more than the <0.5 Ohms we expect in a few feet of small wire. This increases voltage drops, and figuring out what that does becomes one of those brain-buster puzzles.

Hey, it's cheap (cheaper than shotgunning a burnt pedal). Try it.

[Groovenut]

Speaking of deterrence: I think the series protection diode that prevents a circuit from working if you plug in the wrong power supply is pretty good. It forces the user to try to get the right power supply (I suppose they might send the pedal back because "it doesn't work," but at least it's not fried). The problem with just one effect that accepts an incorrect power supply is that it lulls the user into complacency. Then they try to use the supply with another pedal that lacks this capability, or lacks power protection, and at best "it doesn't work" and at worst they let the magic smoke out.

Of course, accepting a wider variety of supplies is technically superior, but in music equipment, no equipment is stand-alone.

My main reason in even entertaining this subject was largely academic. I always use a series schottky in my builds and 90% of the time is fool proof. However, the 10% that plug AC power supplies into DC pedals is why I am going down this rabbit hole. IME, when an AC supply is mistakenly used it kills or moderately damages so many things, it's just better to replace everything. If I could find an inexpensive way to avoid that scenario, I'd implement it. Mind you I would not change the labeling on the pedal. It would still say 9-18VDC center neg

[Groovenut]

> use the lowly 2N7000

ON-resistance is like 5 Ohms. Much more than the <0.5 Ohms we expect in a few feet of small wire. This increases voltage drops, and figuring out what that does becomes one of those brain-buster puzzles.

Hey, it's cheap (cheaper than shotgunning a burnt pedal). Try it.

From the datasheet it appears they want you to shoot for a 30mV drop across each mosfet. In a low draw pedal, you could probably get away with the 2N7000 (30mV/10mA=3Ron) without much issue I am guessing. There isn't a given block diagram for this part so it's hard to tell whats inside. Other similar parts for replacing a single diode I have looked at, use basically 2 comparitors and a charge pump to drive the fet driver. Like you said Paul, it's cheap enough to buy one to experiment with so that may be the next step. Thanks for the input!