New Portable Amp - "Pocket Class A" by xrk971 - now available as complete PCB
Apr 13, 2017 at 9:31 PM Post #151 of 427
Nice setup. Looks like you are mostly an earphone type guy rather than headphones? SE Class A amps are ideal for earphones where ultralow self noise is important because the phones are so sensitive that noise / hiss / hum can be very apparent on an amp that is less than silent.

Did your rechargeables arrive yet?
 
Apr 13, 2017 at 10:26 PM Post #152 of 427
Nice setup. Looks like you are mostly an earphone type guy rather than headphones? SE Class A amps are ideal for earphones where ultralow self noise is important because the phones are so sensitive that noise / hiss / hum can be very apparent on an amp that is less than silent.

Did your rechargeables arrive yet?

 
The rechargeable batteries are still in-transit, so I picked up a pair of 9V Alkaline batteries this afternoon - just in-case I need a new set, before the others arrive.
 
Apr 15, 2017 at 2:25 PM Post #153 of 427
I was looking around to see what sort of discrete circuitry would be required to convert from single-ended to differential (balanced), and the simplest representation consists of 3 resistors, and 2 NPN transistors.
 
Some additional components will be needed, if operating from a single +18VDC supply, rather than a split supply - since if Vee = GND, and you GND the Base input on Q2, the circuit won't work.  I guess the additional components would set the bias voltage on the Bases high enough above GND, to flow enough current through the Collector resistors, so that with no Audio input, the voltage at Vout is ~9VDC.
 
(From: https://en.wikipedia.org/wiki/Differential_amplifier)
 

 
 
Two of these circuits would take the single-ended (L+ / GND) (R+ / GND) input, and produce a balanced (L+ / L-) (R+ / R-) output - which could feed the input to your XRK Class A amp boards.
 
Currently, each of your amp circuit boards has a dual 10K potentiometer, to control L / R volume (by attenuating the input to the amp).  So, if you're using two boards, in a balanced configuration, you have to adjust two separate knobs (one for L, one for R) to control volume.
 
It's a bit tricky trying to determine the best place to connect the potentiometer(s), if the amp can be used as - 
Single-ended Input to Single-ended Output
Single-ended Input to Balanced Output
Balanced Input to Single-ended Output
Balanced Input to Balanced Output
 
Apr 15, 2017 at 2:52 PM Post #154 of 427
Let me look into making a discrete balanced drive converter. What you show is a differential input stage which I try to avoid. Most modern amps have differential transistor inputs but they inherently cancel 2nd order distortion. BJTs also have a finite current flow at the base. If you put a pot in front of this you will get scratchy sounds when changing volume. That's because there is DC current through the pot. Notice that the pot action on my amp is noiseless. That's because of the JFET input stage. JFETs have no current flow through the gate. Perhaps there is a way to make the converter from maybe 6 of the same JFETs.

But if your Onkyo Dap already has balanced output why not just use that? I would just set volume to max on both and have gain set identically in both amps. Volume control would.ve via DAP.
 
Apr 15, 2017 at 3:09 PM Post #155 of 427
Let me look into making a discrete balanced drive converter. What you show is a differential input stage which I try to avoid. Most modern amps have differential transistor inputs but they inherently cancel 2nd order distortion. BJTs also have a finite current flow at the base. If you put a pot in front of this you will get scratchy sounds when changing volume. That's because there is DC current through the pot. Notice that the pot action on my amp is noiseless. That's because of the JFET input stage. JFETs have no current flow through the gate. Perhaps there is a way to make the converter from maybe 6 of the same JFETs.

But if your Onkyo Dap already has balanced output why not just use that? I would just set volume to max on both and have gain set identically in both amps. Volume control would.ve via DAP.

 
There are some articles on audioXpress, originally published in Audio Electronics by Erno Borbely in 1999, where he shows circuits which use JFETs - http://www.audioxpress.com/article/JFETs-The-New-Frontier-Part-2
 
Apr 17, 2017 at 9:50 AM Post #157 of 427
If replacing the two 9V series connected batteries, with one higher capacity Li-Ion battery pack, which would be better for the amp? 
 
14.4V @ 3400mAh (68mm x 74mm x 20mm)  Uses a 16.8V charger, and has 11.2V Over-Discharge Protection.

18V @ 3400mAh (68mm x 90mm x 20mm)  Uses a 21V charger, and has 14V Over-Discharge Protection.
 
I realize that neither will fit in the Altoids tin, so I'm looking for an alternative housing.
 
Thanks!
 
Apr 17, 2017 at 10:04 AM Post #158 of 427
  If replacing the two 9V series connected batteries, with one higher capacity Li-Ion battery pack, which would be better for the amp? 
 
14.4V @ 3400mAh (68mm x 74mm x 20mm)  Uses a 16.8V charger, and has 11.2V Over-Discharge Protection.
 
18V @ 3400mAh (68mm x 90mm x 20mm)  Uses a 21V charger, and has 14V Over-Discharge Protection.
 
I realize that neither will fit in the Altoids tin, so I'm looking for an alternative housing.
 
Thanks!

The 18v one will give you better performance for sure - just make sure your case allows adequate cooling as the MOSFETs will each be disspating about 0.7W each and the resistor arrays 0.6W each or 2.6W total heat generation vs 1.8w total output when running 16v.  You might want to add a 0.47R 1W series resistor between the batteryand the amp to limit the in-rush current that may trip the battery to go into self-protect when turning on.  Try without it first of course.
 
Apr 17, 2017 at 11:24 PM Post #159 of 427
I finally called eBay and asked them what's the deal with making me wait a month to be able to list another item - got it all straightened out so I am now back on eBay (as well as Etsy).  So if you have a preference for eBay you can go there.
 
Here is the NHB Edition listing:
 
http://www.ebay.com/itm/-/222479147739?
 
Here is the standard one:
 
http://www.ebay.com/itm/222479138123?
 
or if you prefer Etsy - options for bare PCB's etc are available there:
 
https://www.etsy.com/shop/XRKAudio?ref=hdr_shop_menu
 
Apr 17, 2017 at 11:46 PM Post #160 of 427
I was looking around to see what sort of discrete circuitry would be required to convert from single-ended to differential (balanced), and the simplest representation consists of 3 resistors, and 2 NPN transistors.

Some additional components will be needed, if operating from a single +18VDC supply, rather than a split supply - since if Vee = GND, and you GND the Base input on Q2, the circuit won't work.  I guess the additional components would set the bias voltage on the Bases high enough above GND, to flow enough current through the Collector resistors, so that with no Audio input, the voltage at Vout is ~9VDC.

(From: https://en.wikipedia.org/wiki/Differential_amplifier)





Two of these circuits would take the single-ended (L+ / GND) (R+ / GND) input, and produce a balanced (L+ / L-) (R+ / R-) output - which could feed the input to your XRK Class A amp boards.

Currently, each of your amp circuit boards has a dual 10K potentiometer, to control L / R volume (by attenuating the input to the amp).  So, if you're using two boards, in a balanced configuration, you have to adjust two separate knobs (one for L, one for R) to control volume.

It's a bit tricky trying to determine the best place to connect the potentiometer(s), if the amp can be used as - 
Single-ended Input to Single-ended Output
Single-ended Input to Balanced Output
Balanced Input to Single-ended Output
Balanced Input to Balanced Output


I'm looking into this option for a discrete Class A balanced line driver that will convert single ended input into balanced phases for driving the power amp portion in balanced drive mode. It will probably add 6 more JFETs but have the advantage that the signal will only be processed with a SE Class A zero feedback topology.
 
Apr 18, 2017 at 11:18 AM Post #161 of 427
Just found the same set, on Amazon, for $96 - https://www.amazon.com/Hitech-Four-Lithium-Polymer-Batteries-Charger/dp/B00890HIE4
And they've got the 2 batteries, plus charger set, for $55 - https://www.amazon.com/Hitech-Two-Lithium-Polymer-Batteries-Charger/dp/B0088R79RE

On Amazon, there's more info about the battery - and it says the nominal voltage is only 7.4V.


I can report that the Hitech 9V @ 720mAh Li-Ion batteries DO NOT work in the Pocket Class A NHB amp.
When the knob is turned from Off to On, there's a soft "pop" in the earphones, and the LED flashes on, then fades off.
It's probably the in-rush current causing a battery protection circuit to activate.
If the earphones are disconnected, and the power is switched On / Off / On rapidly, the LED lights up brighter, but the amp doesn't stay on.

So for now, I'll just have to buy a box of Alkaline batteries.
 
Apr 18, 2017 at 11:34 AM Post #162 of 427
Sorry to hear that Dbaldock9,  I know you have been waiting for those for some time now - I hope you can return them.  You might want to try the EBL 9v Li-ion ones.  The do suffer in-rush the occasional self-protect shutoff but can be made to work reliably.  Mine actually do not suffer any auto shutdown now. It seems the more you use them, the less this happens.
 
Here is what I use:
 
https://www.amazon.com/EBL-Self-Discharge-Lithium-ion-Rechargeable-Batteries/dp/B00EQ3U2AA
 
Apr 18, 2017 at 12:00 PM Post #163 of 427
  Sorry to hear that Dbaldock9,  I know you have been waiting for those for some time now - I hope you can return them.  You might want to try the EBL 9v Li-ion ones.  The do suffer in-rush the occasional self-protect shutoff but can be made to work reliably.  Mine actually do not suffer any auto shutdown now. It seems the more you use them, the less this happens.
 
Here is what I use:
 
https://www.amazon.com/EBL-Self-Discharge-Lithium-ion-Rechargeable-Batteries/dp/B00EQ3U2AA

 
I wonder how small of a series resistance could be used to keep the self-protect circuit from tripping?
 
Apr 18, 2017 at 12:30 PM Post #164 of 427
Try a 2.2ohms 0.5W resistor, then go lower, double it in parallel for equivalent 1.1ohm - the smaller, the better for battery life. Or go the other way if 2.2ohms doesn't help. If you did not want to suffer more than say, 0.5v drop across resistor and want to allow it to pass 200mA then Ohm's law (V=IR) gives you R=V/I=0.5v/0.2A)=2.5ohms. Which is why I picked 2.2ohms.
 
Apr 18, 2017 at 8:13 PM Post #165 of 427
Here's an interesting Single to differential driver. It might even work with two JFETs on the output.



I want to avoid IC opamps but if I used one, a THAT 1646 would probably be the IC of choice. it would be simple and very transparent - ultra low distortion and the SE amps can then do their magic.

https://cdn.sparkfun.com/datasheets/BreakoutBoards/THAT_1606-1646_Datasheet.pdf
 

Users who are viewing this thread

Back
Top