stax mafia electrostatic portable

[JoaMat]

Thanks a lot for this project, Kevin and Kerry.

At least the amp sections seem doable , even with my old eyes and clumsy hands.

Is the board two layers or worse?

[judo]

When you say this I always feel myself f*ing old. I am the older and I did not see but Kevin is somewhere near too I think. Most of us would like your trembling hands🙂  I would try to do it myself but it seems too hard. Only the half of  the parts it has not much meaning.

[sweetleaf]

Hello people. 

This is a stunning little amp, amazing development by the team again. Kerry soldering looks good 

Can I ask if there is XMOS for the USB?

Certainly if this was offered as DIY, the PCB's would need to have difficult IC's part populated by the board house.

[Kerry]

The board is two layers, but I had to use .5mm spacing for the USB. It also has a lot of 0402 components. I don’t think you could use a mill for the PD section. Also those USB and battery management chips were a nightmare.

It’s funny that after placing all those 0402 parts the 0603 stuff seems easy and the 0805 parts seem gigantic.  

I used an oven to bake it since I was afraid the hot air station would blow those smaller parts away. 

The DAC will use the XMOS input and will be on a separate board. We’ll have a version of the input board that is analog only. 

[Kerry]

Some more good progress.  The amp is up and running  

I've got two issues on the board, but the amp, power supplies and USB-C PD / battery management are all working great.

The wires going to the pot are temporary.  There'll be a board for the analog/DAC inputs.  The PD chips are small with lots of pins on them.  I'm surprised I could solder them  

 

Lots of 0402 chips on the right side of the board...

 

Here's the amp running with the battery (I need to have the input board done before I can put the batteries in the chassis).  This is with the amp running and USB-C charger plugged in.  Below, it is sending .18A to the batteries.  The battery management will stop charging once the batteries are at about 12.5V.  It drops the current as it gets closer to a full charge. 

 

With the charger plugged in and batteries fully charged, it doesn't draw any power from the batteries.  

At the moment, it negotiated 5V from the charger (source) and is boosting it to just above the battery voltage.  If we wanted, we could source power so it could charge you phone too.

I'll do a drawdown test tomorrow to see how long it will run on a charge.

Overall, I'm very please with this amp.  Kevin's amp design is fantastic.  I've been listening to it for the last several hours with my SR-007's  

 

[luvdunhill]

13 hours ago, Kerry said:

Some more good progress.  The amp is up and running  

This has been a handy device for me - even has pads to control it from a micro controller:

https://www.amazon.com/DROK-Adjustable-Electronic-Discharge-Resistance/dp/B07KQS74C6/ref=mp_s_a_1_16?dchild=1&keywords=programmable+battery+load+tester&qid=1619718835&sr=8-16

Edited April 29, 2021 by luvdunhill

[Kerry]

Thanks.  I think I'll pick one up.

I've had the amp playing from 9:00 this morning (did I mention it really sounds great ) and batteries were down to 9.8V at 3:20PM.  They are not perfectly balanced so one battery is at 3.1V which is below the minimum 3.3V recommended.  So looks like I'm getting a little over six hours with these batteries (3x2000mAH).  I can look for larger batteries and also play with some optimizations.  I've also been running additional 5V and 3.3V supplies for the DAC that consume extra power.  They're not needed for just the analog input. 

The balancer should have kicked in to turn off the battery pack, but it didn't.  I suppose it would have at some point, but I plugged it back in to charge.  I have a different and smaller balancer that I'm going to try.  Maybe that will do a better job.  

[JoaMat]

Done some soldering exercise today. This small board contains 100 x 0402 10K resistors.

Hand soldered with a Weller station, XNT-H tip and 0.3 mm solder wire. A few of the resistors are not flat to the board but I think they are acceptable soldered. 

Looking at your board, Kerry, there seems to be some really nasty things. Any chance for part numbers? I have to convince myself that the best way is to pay for a prepopulated board…

Edited May 5, 2021 by JoaMat

[Kerry]

You’re hired 😄

[Kerry]

I just finished designing this battery protector board.  It has a built in balancer to try and keep the cells at the same voltage.  I added a thermistor that gets fed up to the battery management on the amp board.  This also has it's own thermistor in the middle of the board.  I'll set this to about 50 C for both and do some testing once I get this built.

The board is 10mm x 56mm.  

[luvdunhill]

1 hour ago, Kerry said:

I just finished designing this battery protector board.  It has a built in balancer to try and keep the cells at the same voltage.  I added a thermistor that gets fed up to the battery management on the amp board.  This also has it's own thermistor in the middle of the board.  I'll set this to about 50 C for both and do some testing once I get this built.

The board is 10mm x 56mm.  

Awesome! I tried a “remote” thermistor (Murata NCP18XH103) and ended up just using a LM234 and some resistors. 

[Kerry]

The TI chips I'm using for the USB-C and protection/balancer already have a mechanism built in, so I just followed their recommendations.  I found some 0603 parts (RT1, RT2 one the board).  They are 10K resistance to start and about 4.3K at 50 C.

I found new batteries that are 2400mAH that should give a little over 8 hours playing time.  Over voltage is 4.2 and minimum discharge voltage is 2.75V.  I'm using 4.2V for OV and 3.0V for UV.

[Kerry]

On 5/5/2021 at 11:44 AM, JoaMat said:

Looking at your board, Kerry, there seems to be some really nasty things. Any chance for part numbers? I have to convince myself that the best way is to pay for a prepopulated board…

The USB-C and battery management chips are crazy insane.  It would not be possible to solder with an iron.  They both have one or more exposed pads underneath the chips.

Here's the USB-C PD controller... 

Those giant looking resistors on the left of the chip are 0402

Edited May 9, 2021 by Kerry

[JoaMat]

Looks very nasty... really nasty.

[judo]

It seems as the PCB fabrication the PCB population is out of an average diy-er hand. Now microscope, oven, stencil fabrication etc. we have to buy or make? robots in the near future. I do not know this is good or bad. Soldering was the part of the game but will not be missing much, I think.

[Kerry]

Here's the final board I just sent out to get manufactured...

 

I added some more optional parts that help stabilize the current sense circuitry.  The board is now 10mm x 58.4mm x 1mm.

Parts are ordered, so it's a race to see if Mouser can ship my order before the boards arrive.

[Kerry]

In other news, I just got back the analog input board and assembled the amp.

 

The sandwich is 6.2mm thick + bottom components for a total of about 9mm thick.  Batteries are about 10mm on top of the input board.

 

Back plate on and plugged into USB-C (charging LED on)

 

Why we did this... 

Portable music  

Just analog from my phone at the moment.  DAC is up next

 

PS  The Weight is 12oz or 340g

Edited May 10, 2021 by Kerry

[tkam]

I love how absurd the stax plug looks on such a tiny amp

[s2030081]

On 4/26/2021 at 3:24 AM, kevin gilmore said:

makes no sense to make this a diy kit. the main board is more than 200 components on both sides of the board. The board house wants about $60 to assemble this board.

no one i know can hand stuff that board for $60. The dac board is likely going to use the ess chip which has nda agreements associated with it. So no gerber files or schematics either.  Still should be much cheaper than the d10 or kse1500.

If this will be a product to sell, could there be a version for Shure kse? and also a version of analog in only?

When I use the SRM D10, I will use the analog in only most of the time for saving battery.

People always say the KSA amp from Shure was bad. I am interested how far my kse iem can go 😁

Plus the battery in KSA amp die every year from my experience...

[kevin gilmore]

modifications would be required to lower the bias to 200v from 560.

it is unknown whether the headphones can take the extra voltage swing of this amp.

one version of this amp will definitely be analog input only.

lithium batteries suck. get used to it.

Edited May 15, 2021 by kevin gilmore

[s2030081]

Quote

one version of this amp will definitely be analog input only.

At least some good news here🥰 Thanks!

[Kerry]

I got the board and parts in for the battery protection/balancer and was able to get one built. 

I’ll try to get some testing in tomorrow. 

[nzastro]

Wow! You guys ROCK. Awsome! I'm an old newbie as far as headphones go. Never liked headphone sound until trying the Stax last year. Now I'm in the middle of building a 450v KGSSHV for my L700 and dreaming of the 009mk2 or 007mk2?

I will Definately be a starter for your portable amp. 

Due to several health provider slip ups I now have permanent health issues so headphones are more suited. 

Pity I live in New Zealand so can't sue them. If I lived in the States I could have sued and would be able to afford the best system🤪. 

Actually I'm very happy to live here. New Zealand is an awsome country and fingers crossed no Covid at present. 

Seeing your progress with this amp really gives me something to look forward to. 

Well done. Keep safe n well. 

All the very best from down under. 

Cheers! Tony. 

 

[Kerry]

Initial testing on the balancer is looking good. I’ll need more testing to confirm, but the batteries are much closer than before. This should help to extend battery life. 
 

[Tachikoma]

On 5/15/2021 at 9:04 PM, kevin gilmore said:

modifications would be required to lower the bias to 200v from 560.

I suppose it wouldn't be possible to make the bias voltage switchable?