AA HPA V. 1.0 Info

[morphsci]

I just unpacked a box from our last move in 2005. I found my HPA V. 1.0 but no power supply. When I packed it to move it was acting up so I just want to get workable power supply to see if it actually works. I believe the basic power supply for the HPA was 12V and 800 mA. Can someone confirm or refute this. Also if you know the dimensions of the power supply tip that would be helpful. Thanks.

[jinp6301]

Hydrocity has a HPA i think

you should ask hydro

[hYdrociTy]

The stock power supply was a piece of shit on any standard. Some retard decided to use a stereo 1/8 plug as a power connector carrying dual supply. Apparently they didn't think people would attach or detach the power supply while it was plugged in the wall. It didn't help that there were no on/off switches anywhere either

The basic supply is a +/- 12v dual psu probably around 500ma. I suggest at least 1amp and you can safely go up to +/- 14v on this thing since there was an upgrade psu for it that was +/-14v and 1amp.

I would say a nice power supply is two treads and an amveco 62063 transformer. Then you can dial the system from +/-12 to 14v depending on your taste and heatsinking blah blah.

the plug is configured as so:

tip: positive

ring: negative

sleeve: ground

This tiny amp kicks plenty of ass with a nice power supply and has loads of power ~1 watt output with 1amp psu. Sounds great with K1000. The little amp chassis gets hot as hell though.

here is a temporary bootleg rig i built for it.

Still slowly saving up for a hammond and some treads...

[morphsci]

You are da' man. Thanks, that is exactly what I needed to know.

[luvdunhill]

hey, where did you buy that transformer, just out of curiosity?

[jinp6301]

man someone at AA was an idiot to choose an 1/8 plug as the power connection

[JBLoudG20]

man someone at AA was an idiot to choose an 1/8 plug as the power connection

I think hydro did that.

[jinp6301]

The stock power supply was a piece of shit on any standard. Some retard decided to use a stereo 1/8 plug as a power connector carrying dual supply. Apparently they didn't think people would attach or detach the power supply while it was plugged in the wall. It didn't help that there were no on/off switches anywhere either

seems like it AA decided to make it 1/8th?

[JBLoudG20]

seems like it AA decided to make it 1/8th?

Oh oops. Missed that one.

[hYdrociTy]

yay i has a two treads. this thing now draws 13 watts. the transformer and iec/switch blah blah is behind the screen out of sight, sandwiched between 2 sheets of ers paper. Bass is much more apparent vs the other power supply, and it now takes up to like 3.5v input without clipping. It is rated 3.7 input to match the dde3 but somehow it just doesn't do it. Maybe when i get a 25va transformer and up the voltage to +-14v it will finally fully take in the disgustingly high output of the dde.

It drives K1000's well with the bootlegger psu before so I have lots of hope. I bet the day I power this with a sigma22 I'll finally have a nice compact amp for my future earspeakers...

[Pars]

Mainly addressed to hydro, but anyone else chime in. I'm working on morphsci's HPA1.0 now and I see that it has 2 sets of 12V regulators (7812 / 7912). Since it regulates the PSU input on board, it would seem that using something like a sigma22 on this would be wasted? I definitely think it needs current, but perhaps a raw dual DC supply capable of 1A on +/- might be better? I haven't traced it out to see why they are using 2 sets of regs however... (TO220s on right side of pic)

[hYdrociTy]

Thanks for revealing the mystery inside

I could never find any hex tool that small

Yea I see what you mean now...They have been known to do things that make absolutely no sense, so I wasn't surprised with all the excess crap in the power supply. So with their own upgraded power supplies there would be six or maybe eight regulators.

Well good luck on that unit. Maybe you could re cap that thing, it runs hotter than you can possibly imagine, so the components inside are probably on their wane for being more than 12 years old....

A better power connection would be nice, say a mini-din.

[Pars]

Since I've never seen any of their power supplies, upgraded or not, what can you tell me about them? The input voltage must be higher than ~14.5V for the onboard regulators to function (if I'm reading National's datasheet correctly). I don't know how this might respond to a sigma22 configured for +/-12V... since I have one, I might try it .

[n_maher]

But would it be so bad for the on-board regs not to work if you were already feeding it a well regulated, presumably low noise supply?

[hYdrociTy]

I for one don't think it's a waste to have a good power supply even if the thing has a bunch of mediocre regs. I mean regs don't add noise right? *clueless* Maybe you could like... straighten the components while you're in there. I've never seen so much bent stuff...

[Pars]

But would it be so bad for the on-board regs not to work if you were already feeding it a well regulated, presumably low noise supply?

I would think that would be fine, but had to go read the datasheet and look at the pictures (ok, graphs ) to see how regs behaved if they were fed too low of voltage for them to regulate. I haven't played around with vregs/PSUs enough to know that. It seems that as long as the input voltage is above a certain threshold (looked like ~2V for a LM340 5V reg.) that it just passed the input voltage thru until it was high enough (7.5V) to regulate.

[morphsci]

Feel free to experiment with the unit as you see fit. I'm O.K. taking any risks with that as I see figuring out how this puppy works is the most important objective.

[luvdunhill]

I have a DAC I'm working on that has something like 8 regs per channel. There are three rather simple unregulated supplies feeding them, very similar to the unregulated section of a STEPS. I also experimented with a pair of sigma22 and a sigma11 (yes, crazy) and bypassing the regs and I couldn't hear any difference, in fact the regs sounded better to me depending on how little sleep I had at the moment. I was planning on trying a "TeddyReg" design, or even other similar mini regulators and just haven't gotten around to it. Problem with those, is they are hard to heat sink. I think the key here having just enough voltage head room, so that most of the heat is from current consumption. I changed all regs to be LDOs and thought it was a positive change, and looks to me to be a good choice in this situation as well.

Then again, take this with a grain of salt. I'm using ceramic capacitors as bypasses right next to all the DAC chip sockets, so what do I know

[Pars]

I hadn't really intended on experimenting on the unit since it isn't mine, and the PCB appears to be a bit fragile in terms of trace and thru-hole plating lifting. I had simply told morphsci that I would take a look at it and see if I could get it working, which I did.

He had mentioned that he doesn't have a power supply so in a PM I had somewhat offered to look into it if he wished and see what something suitable would cost. Initially I had thought that a pair of Treads or a sigma22 would be nice, or even a +/- 12V Elpac I have laying around. I would prefer to leave everything else stock, and not get involved with changing out the 3.5mm mini power jack, caps, etc.

[luvdunhill]

oh I see. I wouldn't go with anything more than an Elpac... I mean, the money for a sigma22 could much be better spent on, say tossing the amp and building another one.. i mean, improving the circuit (somehow)...

Looks like opamps for voltage gain and then solid-state current gain into class A? solid-state output stage. Also, lots of 'styrene.. I like that.

edit (again): why not try battery power?

[Pars]

Are batteries any cleaner than a sigma22 or one of Gilmore's PSUs? My guess would be no... IIRC in the original headwize article for the Gilmore Dynamic (dynalo), Kevin stated that the PSU was cleaner than batteries. Plus it requires +/- power input so you would have to use 2 batteries or a rail splitter.

[luvdunhill]

I'd bet the batteries + reg is just as clean. I don't know if the supply Kevin was referring to in the comparison used a regulator after the battery, but that's a good question. I thought that there was an AA commercial battery power upgrade for these units?

edit: what's the current draw?

[Pars]

I'll need to more accurately measure the current draw, but the bench PSU said ~>200mA for the positive and ~350mA for the negative rails. Not sure why the neg seemed to be drawing moar power.

Bench PSU is a relatively cheap B&K, don't know how (in)accurate the meters are.

[luvdunhill]

I'll need to more accurately measure the current draw, but the bench PSU said ~>200mA for the positive and ~350mA for the negative rails. Not sure why the neg seemed to be drawing moar power.

Bench PSU is a relatively cheap B&K, don't know how (in)accurate the meters are.

that's not all that uncommon. Looks like a CCS before the opamp, that could be doing it.

[Pars]

Got a chance to play around with this more today. First thing I did was change out the power entry caps (470u 16V) for a pair of Pana FM 470u 25V. The 35V version of this cap won't fit BTW.

The caps on the reg outputs are Nichicon Muse FX (I think) 220u 25V, so I left these alone as IIRC 3 terminal regs don't like low ESR on their outputs. I thought the input caps were worth changing on their voltage (16V, a bit low for a 14.5V + supply input) and their temp rating (85C).

I have been doing listening tests using a Sony PCDP line out into the HPA and then into my AKG K601s. I tried the following:

• +/- 12V bench PSU
  
• +/- 15V bench PSU
  
• +/- 12V Sigma 22
  
• +/- 15V Sigma 22
  

I really didn't notice much if any difference between these. The 12V levels gave working DC of ~10V at V+/V- (pins 8 and 4) of the OP275s as it is too low for the regs to regulate. The 15V levels gave ~12V at V+/V-. I was using an Enya CD as test material, and did notice a bit of upper midrange "steeliness", but this could very well be the source.

Power draw was backwards from what I previously had posted: + rail ~350mA; - rail ~260mA.

At this point, I would think a pair of Treads might be the best and most cost effective PSU to go with. The Sigma22, while nice (really nice), doesn't seem to buy you enough to warrant the cost. Which of course wouldn't keep me from using it

None of the Elpacs put out enough current to consider, and they have gone up considerably in price (the WM071 used for the GLite is now $43 from Newark... it is rated at 240mA per rail).