O2 AMP + ODAC - Page 26

lol, yes of course. I'm at work in multiple meetings today so I'm unable to check it more thoroughly at the moment. A buddy check the PS for output and found nothing. Now, where to get a replacement...asap. I don't want to be without it for long. :p
 

 

Just in case! A friend of mine, working on a big, expensive equipment-support contract, once drove 2½ hours into the mountains on a service call. When he arrived, he just plugged the transmitter into the electrical outlet, wrote up a invoice for 5 hours' service, and returned to the big city. It's the kind of mistake I would make (but haven't yet.)

 

Near the bottom of Objective2+ObjectiveDAC Combo Operating Instructions, there is a list of power supplies recommended by the designer. They were available from Mouser a week or so ago, and some are (or were) carried by MarkerTek; DigiKey is another major supplier. You may have local electronic supply shops.

Ah, Thanks for the link. Didn't know that was there! Appreciate the response.

 

Edit: It's been plugged in for 30 minutes and as cold as a brass doorknob. I'm going to try and find one locally so I don't have to go without over the loooong holiday weekend.

Edited by Agavehound - 11/20/12 at 12:00pm

Hey gang, I've located an AC adapter locally, but I don't know if the mA rating is too high. Ouput is variable @9-13VAC and 800mA. Is that too high and will it damage the 02/ODAC unit?

 

That will work just fine - a current rating that's too low for a device is what can cause problems.

 

The current rating (mA) means that the adapter is able to supply up to 800mA of current, but it's the device you plug in that determines how much current is drawn. The O2 instructions (http://www.jdslabs.com/pdf/Instructions_O2.pdf) specify a minimum of 200mA, meaning that the O2 shouldn't draw more than that from a power adapter.

 

Also, AC adapters that just do one output voltage are usually cheaper, AFAIK.

Excellent! I got the variable adapter since it was the only one I could find locally and easily. Hopefully it will last at least 2x as long. (since it was 2x the price. lol)

13V is a little low, may have problems.

 

In reality, it depends on how low your home line voltage is, the kind of load you put on the amp (max volume or bench testing into lower-impedance loads requires more power draw).  Voltage you get out is more-or-less just a fixed fraction of the voltage from the wall, so if the line voltage is lower, then the output voltage will be lower too.  Anyway, if it's rated to be 13V at that setting, it might be sitting higher than that at the loads you'd run, considering that you won't pull anywhere close to the 900 mA it's rated for.

 

It's probably okay, but it's probably best to check the voltage with a multimeter before using.

 

Actual recommended input to amp is 14-20V AC.  Default transformer in NA yields more like 13.5V unloaded despite being rated at 12V, so that's mostly okay.  If it were actually 12V AC, that would be an issue.

Edited by mikeaj - 11/21/12 at 2:40pm

 

I believe the COMPLETELY ideal output should be about 14/15V AC and more than 480mA to be absolutely safe. But generally, it's like above. :-)

If that's true, then why are they recommending the 12VAC 200mA supply. That's the one I got and it lasted all of 8 days

 

Because it's the cheapest and it's going to work with the majority of headphones... But if you read carefully through the entire blog, you will find several notes about ideal transformer and why.

 

You want 14-20V and more than 400 (or even 500) mA to be absolutely sure. I have 15V + 650 mA to be totally absolutely sure that I meet everything mentioned on the blog.

Edited by MHOE - 11/21/12 at 4:11pm

 

The designer has recommended more than one A/C transformer. Of the ones named on their instruction sheet, JDS Labs ships the one with the lowest current (200mA), output voltage (12VAC), and Mouser price ($5.55).

 

It's not clear that their choice of models is responsible for an early failure. After all, these are inexpensive, mass-produced parts. I could be wrong about that one but there haven't been many reports of failures.

 

I have both WAU 12-200 and 16-400, without any troubles... Purchased 15-650 with europlug for about 5-6 dollars but still it works very nice. No troubles at all. :-)

Unfortunately in those few instances when "Voldemort" made a mistake he tended to go into defensive mode rather than fix the problem it seemed. That WAU12-200 transformer choice was something else I tried to talk him out of at one point last year after he first posted the O2 design. One problem with current capability of the transformer secondary is shown in the simulation below. This is done for a 16VAC transformer but the problem is the same. A half wave rectifier, like the two in the O2, charge up the filter capacitors with a very short high(er) current spike after the rectifier turns on each cycle, as opposed to a lower current spike with full wave or bridge rectifiers. That is a design trade-off in using a half wave rectifier. Note in that LT Spice simulation a 200mA (rms) continuous load produces a 800mA charging spike in each 470uF filter capacitor, and hence through the transformer secondary.

The chips in the O2 pull about 22mA just as they sit (quiescent current consumed by the chips) and the batteries pull about 22mA of charging current during the maximum part of the charge curve if they are completely dead (7Vdc each where the PM circuit cuts off) so a headphone load of more than only 3mA or so would produce a charging spike of more than 200mA in the transformer.

If the transformer secondary is rated less than the amount of the charging current spike the core will saturate. That can cause heating and also some noise generation. In the case of heating, the Triad wall adapters may contain thermal fuses (can't find any reference to that, but pretty good guess to be UL certified) which may eventually open up if the transformer gets hot enough. My suggestion to "Voldemort" was to specify the 1 Amp WAU16-1000s (Mouser #553-WAU16-1000), for the case when the O2 was fully loaded with 200mA per channel, or the 400mA WAU16-400 (Mouser #553-WAU16-400) for more normal loads that would probably be the case 95% of the time. He eventually added those two under the "optional" column on his BOM. The 400mA transformer would increase the maximum permissible headphone current load before transformer core saturation from 3mA to 56mA.

To summarize, if your headphone load is less than only about 3mA or so, or up to around 25mA if you are not using batteries (so are not charging anything) you would probably be OK with the 200mA transformer secondary rating. The 12Vac transformer secondary voltage rating is another issue. That is 12Vac if your line voltage is at least 110VAC or so. If your line voltage sags less than that the transformer output will drop lower and the voltage regulators won't have the 3.5Vdc or so across them they like for maximum ripple and noise rejection from the datasheet. At one point some folks had found "12Vac" transformers made for LED lighting, in reality outputting 11Vac, causing the O2 regulators to actually go into dropout. 14 or 16Vac would provide more headroom against line voltage fluctuations.

If the incoming line voltage from the wall socket is on the high(er) end, 120Vac - 130Vac in the USA, and is known to be fairly stable over the course of the day, then the "12Vac" transformer will be producing closer to 13Vac - 13.5Vac loaded (14.5Vac+ unloaded) and would be OK from a voltage standpoint.

For more tech details, see the second picture down on the left on this rectifier application note for Hammond transformers, the half wave capacitive input situation [can't post the link, just do a Google search on "Hammond rectifier" and it is the first result]. Hammond recommends de-rating the rms secondary current by a factor of 0.28 in the case of half wave rectifiers. So in the case of the 200mA WAU12-200 transformer, that secondary is only good for (200mA)(0.28) = 56mA (rms) continuous, in the case of capacitor input half wave rectification, to prevent transformer core saturation.

EDIT 1 11/22/2012: In fairness to "Voldemort", since he seems to be AWOL these days for some reason, his defense of the first issue a year ago was that the amount of excess heating and/or noise generation in the transformer from small amounts of transformer core saturation would be within specifications. His defense of the second issue was that if low line voltage results in smaller voltage drops across the regulators, or even if they drop out of regulation completely, the power supply rejection ratio of the amplifier op amps is good enough to not matter even if voltage ripple winds up on the power supply rails. True to some extent (many tech details involved), but my view remains why create those marginal operating conditions in the first place when for $1-$2 more USD, the way transformers are priced, you can buy a 14Vac - 16Vac 400mA - 800mA unit and avoid the issues entirely.

EDIT 2 11/25/2012. Added some more information and cleaned up some grammar. Another issue that probably figured into "Voldemort" leaving that WAU12-200 transformer as the primary BOM specification was cost. I can say from some email exchange a year ago that for some reason he was very adamant about keeping the basic parts cost at $30. He considered raising the parts cost even a couple of dollars a big deal. I think his view was along the lines of keeping the O2 very affordable for DIYers.

Capture.PNG 31k .PNG file
Edited by agdr - 11/25/12 at 2:55pm

IDK, in one of my conversations with him, he recommended a 16V power adapter.

 

Thank you for the analysis. I'm not an EE, but I'll read it and do my best to understand.