Did you see the PS Audio humbuster, seems just like a pair of diode bridges and a few caps for $299. Does anyone know the schematic for this type of circuit which allegedly gets rid of the DC from the AC line?
An externally hosted image should be here but it was not working when we last tested it.
Hmmm, this sounds fishy to me.
Note that their site says: removes nearly 1 volt of possible DC from the AC line (...) in an arrangement that removes the direct coupling at the AC’s zero crossing point.
Note that the voltage drop across a diode is appr. 0.6 volt (1 volt rounded).
My first impression would be that it's essentially 2 diodes in parallel in reverse direction, which would indeed remove nearly 1 volt in either direction, lowering peak-peak amplitude by just over 1 volt, add nasty things at/near the zero-crossing and essentially keep any difference in positive and negative amplitude (the DC voltage, if any) intact...
Of course this could just be my paranoid and suspicious nature
Why not try it and tell us if it helps
Note that their site says: removes nearly 1 volt of possible DC from the AC line (...) in an arrangement that removes the direct coupling at the AC’s zero crossing point.
Note that the voltage drop across a diode is appr. 0.6 volt (1 volt rounded).
My first impression would be that it's essentially 2 diodes in parallel in reverse direction, which would indeed remove nearly 1 volt in either direction, lowering peak-peak amplitude by just over 1 volt, add nasty things at/near the zero-crossing and essentially keep any difference in positive and negative amplitude (the DC voltage, if any) intact...
Of course this could just be my paranoid and suspicious nature
Why not try it and tell us if it helps
Richard C said:
A very informative thread, and another lesson learned. Maybe I should work a little on my suspicious and paranoid nature
Still, $299 is an awful lot of money for a few diodes and some caps
At least DIY will pay off in this case, so why not try and build it and tell us if it helps
Cheers,
Remco
Richard C said:I might be missing something but I still don't see how it removes the offset. It appears to reduce amplitude of both halves of the waveform equally so any imbalance still remains.
Think of it first as only a capacitor in the line, that will remove any DC, right?
Now put two diodes parallel to that (in reverse directions); those are to protect the capacitor (for higher DC voltages), so indeed it only works to remove DC up to 0.6V.
If DC offset is higher extra diodes can be added (in series to the other ones). But I gathered from the thread you posted that small DC voltages are enough to cause problems.
For the humbuster it looks like they used full-wave rectifiers (bridges); so cooling can be added (screwed on) easily.
(A bridge connected with only + & - to either side of the capacitor is essentially 2 sets of 2 diodes in series in the same direction, which is why you need two bridges).
Here's what's inside
Guys, this is my first post here. One of my engineers, Jon Magnussen, told me about the post. I am the CEO of PS Audio and am happy to answer any questions for you.
We are an engineering based company, not a sales and marketing company like a lot of others - but of course we do what sales and marketing it takes to make the company fly.
You're wondering what's inside and have figured out there's a couple of bridges and some caps from the picture. Basically this is correct: there are two 50 amp bridges, 8 caps, a PC board, IEC inlet, AC outlet, 4 piece machined aluminum chassis, and that's about it. In fact, it's one of the simpler product we've ever made.
Why $299 retail? Well obviously if you built this yourself you'd pay a lot less - I guess that applies to any DIY project, eh? The chassis alone costs us $60, then you add the parts, the labor, the line cord, the packaging, the incoming freight, the dealer markup, the certifications for safety, the salary of the purchasing and operations manager, the shop foreman....you get the idea. I can't even sell you an empty box for much less than a couple of hundred dollars retail - and PS Audio is affordable compared to some of the "luxury" brands! It's just life in manufacturing.
Here's how it works. Think of four diodes, two in series going one way and two in series going the other way. Now picture the string of two: one with the cathode at the AC line and the other with the anode at the AC line. This is not far off from the way we do it. We use diode bridges not as they were intended, but as a good way to use series diodes in parallel with heat sinking capabilities (remember that power bridges have a big chunk of metal for heat sinking). But for the moment, ignore that these diodes are in parallel (which they have to be for heat reasons).
There's always a little DC component on the line. It's usually less than a volt. Two diodes in series gives you 1.2 volts. So in either direction, the DC cannot get through the diodes as long as it's less than the 1.2 volts. Now there's all these caps. And, BTW, if you build this yourself be careful. You must use the right caps. They need to be AC rated X and Y caps so you don't start any fires. They are pretty expensive (relatively). The caps are aranged such that at AC zero cross, where the diodes aren't allowing any current to flow, the caps couple the AC line to the load. The other caps clean up the sharp transition of when the diodes conduct.
This arrangement basically stops the DC from getting to the equipment. Under heavy loads, the drop across the diodes naturally generates heat, so in the picture you see the prototype HBAC with an aluminum slug on the diode bridges. In real life, the bridges are tied to a machined pad on the side of the chassis and the chassis acts like a heat sink. The diodes can get pretty danged hot, so if you're building this, be careful - depending on your load.
The assymmetry of the transformer, caused by the core being saturated due to the DC, goes away and the laminations rattle less. It's pretty simple. Took us a long time to figure it out, but it's simple. Most things are simple once you figure them out.
There are plenty of folks out there not as adept at circuitry as you guys - so go easy on them. They want an affordable solution to their hum problems, all dressed up real nice like in a plug-and-play manner. The Humbuster provides exactly that and considering what it costs to build and run a company and sell through dealers, it's a fair price.
I am available to answer engineering type questions on most any subject if you like. Feel free to email me. I've been designing stereo equipment for 30 years. I am not an engineer, just a "school of hard knocks" seat-of-the-pants trained guy. Our engineering team has lots of trained engineers - they keep me straight. Well, maybe.
Anyway, have fun! I envy you guys - you have all the fun of audio without worrying about how to make the next payroll.
Guys, this is my first post here. One of my engineers, Jon Magnussen, told me about the post. I am the CEO of PS Audio and am happy to answer any questions for you.
We are an engineering based company, not a sales and marketing company like a lot of others - but of course we do what sales and marketing it takes to make the company fly.
You're wondering what's inside and have figured out there's a couple of bridges and some caps from the picture. Basically this is correct: there are two 50 amp bridges, 8 caps, a PC board, IEC inlet, AC outlet, 4 piece machined aluminum chassis, and that's about it. In fact, it's one of the simpler product we've ever made.
Why $299 retail? Well obviously if you built this yourself you'd pay a lot less - I guess that applies to any DIY project, eh? The chassis alone costs us $60, then you add the parts, the labor, the line cord, the packaging, the incoming freight, the dealer markup, the certifications for safety, the salary of the purchasing and operations manager, the shop foreman....you get the idea. I can't even sell you an empty box for much less than a couple of hundred dollars retail - and PS Audio is affordable compared to some of the "luxury" brands! It's just life in manufacturing.
Here's how it works. Think of four diodes, two in series going one way and two in series going the other way. Now picture the string of two: one with the cathode at the AC line and the other with the anode at the AC line. This is not far off from the way we do it. We use diode bridges not as they were intended, but as a good way to use series diodes in parallel with heat sinking capabilities (remember that power bridges have a big chunk of metal for heat sinking). But for the moment, ignore that these diodes are in parallel (which they have to be for heat reasons).
There's always a little DC component on the line. It's usually less than a volt. Two diodes in series gives you 1.2 volts. So in either direction, the DC cannot get through the diodes as long as it's less than the 1.2 volts. Now there's all these caps. And, BTW, if you build this yourself be careful. You must use the right caps. They need to be AC rated X and Y caps so you don't start any fires. They are pretty expensive (relatively). The caps are aranged such that at AC zero cross, where the diodes aren't allowing any current to flow, the caps couple the AC line to the load. The other caps clean up the sharp transition of when the diodes conduct.
This arrangement basically stops the DC from getting to the equipment. Under heavy loads, the drop across the diodes naturally generates heat, so in the picture you see the prototype HBAC with an aluminum slug on the diode bridges. In real life, the bridges are tied to a machined pad on the side of the chassis and the chassis acts like a heat sink. The diodes can get pretty danged hot, so if you're building this, be careful - depending on your load.
The assymmetry of the transformer, caused by the core being saturated due to the DC, goes away and the laminations rattle less. It's pretty simple. Took us a long time to figure it out, but it's simple. Most things are simple once you figure them out.
There are plenty of folks out there not as adept at circuitry as you guys - so go easy on them. They want an affordable solution to their hum problems, all dressed up real nice like in a plug-and-play manner. The Humbuster provides exactly that and considering what it costs to build and run a company and sell through dealers, it's a fair price.
I am available to answer engineering type questions on most any subject if you like. Feel free to email me. I've been designing stereo equipment for 30 years. I am not an engineer, just a "school of hard knocks" seat-of-the-pants trained guy. Our engineering team has lots of trained engineers - they keep me straight. Well, maybe.
Anyway, have fun! I envy you guys - you have all the fun of audio without worrying about how to make the next payroll.
BTW, one last note. I saw in an earlier thread that my old friend Nelson Pass suggested using to electrolytics in series to block the DC. He's right, this does work, but unless you have a bucketful of capacitors, when you have a heavy current load, like that of an amplifier, the output waveform from the caps is pretty distorted.
So, I don't recommend this from a purists standpoint. On the other hand, if you're only blocking DC to a preamp or CD player, something with a light load, then it's just fine.
So, I don't recommend this from a purists standpoint. On the other hand, if you're only blocking DC to a preamp or CD player, something with a light load, then it's just fine.
I just ran a little simulation. Two diodes in parallel anode to cathode and bypassed by a capacitor. I connected this network in series with a resistive load and fed it with a 5V peak sine wave with a +0.5V offset and..... the offset was still present at the load the only difference being the waveform was corrupted by the diodes. Which bit am I not understanding
Re: Here's what's inside
I've read throuh the earlier thread which seems to conclude (correctly IMO) that this type of circuit will only work provided the diodes never conduct.
The above explanation states that the diodes in the Humbuster do conduct and the capacitors only provide coupling around zero-crossing and judging by the size of them I would expect this to be the case once a load is connected. All the other DC blockers described use much larger electrolytic caps.
Is there any test data to prove that the Humbuster does what it claims? The description suggests that their is an actual DC voltage that somehow exists independently of the AC.
paulmc said:
I've read throuh the earlier thread which seems to conclude (correctly IMO) that this type of circuit will only work provided the diodes never conduct.
The above explanation states that the diodes in the Humbuster do conduct and the capacitors only provide coupling around zero-crossing and judging by the size of them I would expect this to be the case once a load is connected. All the other DC blockers described use much larger electrolytic caps.
Is there any test data to prove that the Humbuster does what it claims? The description suggests that their is an actual DC voltage that somehow exists independently of the AC.
You are correct. You don't want the diodes conducting until after the 1.2 volts. Thus, the DC does not get through if it is less than 1 volt. Then, in the positive half of the cycle (for instance) the diodes conduct until the sine wave drops below their conducting voltage, where the caps take over.
If you have a good meter, like a Fluke, place it on the AC line in the DC mode. Most lines have a DC component and it's pretty easy to measure.
If you have a good meter, like a Fluke, place it on the AC line in the DC mode. Most lines have a DC component and it's pretty easy to measure.
It appears to me that when the diodes conduct all that happens is that equal amounts are trimmed off the amplitude of both positive and negative halves of the waveform. How does this address the asymmetry that leads to the net DC offset?
If the Humbuster has a threshold of 1V and you pass a 2V sine wave with a +1V offset through it, is the output more or less symmetrical than the input?
If the Humbuster has a threshold of 1V and you pass a 2V sine wave with a +1V offset through it, is the output more or less symmetrical than the input?
millwood said:
I agree, the purpose of the diodes should be to protect the caps and if the diodes conduct then the AC coupling is lost and any offset at the input will appear at the output. The diodes impose a limit on how much offset can be removed by defining the maximum voltage across the caps.
The humbuster uses very small caps and the expanation of its operation states that the diodes conduct most of the time. Since the offset affects the whole of the waveform AC coupling for a volt or two at zero-crossing seems pointless.
Anyway, it should be easy enough to demonstrate its effectiveness but I don't see any specs or tests on the PSA website.
If you put 1 volt into two diodes in series and a load resistor at the other end, what do you get out?
Nothing. That's one of the things the diodes do. The caps keep it coupled at low voltages in the AC sense. Remember, on a 120 volt or 220 volt sine wave, there's not much current being drawn at and around zero cross.
Here's one of the ways we test the Humbuster. We take our AC regenerator, the Power Plant. We place a DUT at the output of the Power Plant and measure the audible noise of the transformer. very low as the Power Plant has less than 5mv DC.
We then use the offset rim adjust pot inside the PP to move the DC up to several hundred mv. The DUT transformer hums like crazy - you can hear it across the room. We then insert the HBAC and under the same conditions, the hum vanishes.
It really does work.
Nothing. That's one of the things the diodes do. The caps keep it coupled at low voltages in the AC sense. Remember, on a 120 volt or 220 volt sine wave, there's not much current being drawn at and around zero cross.
Here's one of the ways we test the Humbuster. We take our AC regenerator, the Power Plant. We place a DUT at the output of the Power Plant and measure the audible noise of the transformer. very low as the Power Plant has less than 5mv DC.
We then use the offset rim adjust pot inside the PP to move the DC up to several hundred mv. The DUT transformer hums like crazy - you can hear it across the room. We then insert the HBAC and under the same conditions, the hum vanishes.
It really does work.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.