OPA1688 Super CMOY, 2x 9V with real ground and headphone relay - PCBs

No, two independent paths of USB 5v to DC-DC step up making say +9v each. Now tie +ve of one to -ve of the other. Then make that the new virtual GND. Now you Jane a +9/-9v supply with GND referenced to common of both power supplies.

Am I getting warmer? Apologies for the newbie skills. :D

An externally hosted image should be here but it was not working when we last tested it.
 
Here is an easy way to convert a Super CMOY to desktop use if anyone should be interested, Mouser #919-RW-0509D:

RW-0509D RECOM Power | Mouser

That is an isolated +/- 9Vdc & 167mA per rail output, DC-DC converter (4.5V to 9V input) that about 10mm smaller in each of the three dimensions than the size of one 9V battery. :) Could be fed by regular USB 2.0 (500mA) thanks to the 73% efficiency, or USB 3.0 (1A). With the 167mA out on each rail it is enough to power either a single (75mA per channel) or double chip (150mA per channel) Super CMOY. The part is $18. Mouser has cheaper ($8) DC-DC converters that are 4.5V to 5.5V in and 111mA out, but they run at 45KHz. They are also SIPs that are about the same height as the 3.5mm jacks on the board. This one runs at 100KHz to 150KHz, out of the audio band, and is low profile DIP.

To DIY it just flip the convertor over and glue the top of the part to the bottom of the mint tin with RTV or hot glue. Extra credit: use a thermally conductive RTV if you can find one at a reasonable price, which will heat sink it to the mint tin. Then let the pins stick through some perfboard and point to point wire it. Cut off the battery snaps and solder those wires to the output pins of the converter. Then the power wires from a USB cable through a hole in the tin to the converter input, plus 499R load resistors across each converter output.

USB 2.0 A Type Male Plug 4 Pin 4 Wire Data Charge Cable Cord Connectors DIY 30cm | eBay

With that cable just solder the two power wires to the converter pins and away it goes!

@agdr Do I have the connections and resistors right for what you have described above?

An externally hosted image should be here but it was not working when we last tested it.
 
Why 4 USB cables? Just one per DC step up.
Slow burn here, understand now.

What would be a recommended voltage setting for both of the MT3608 DC DC Step Ups... 9V? Or would a higher or lower value between 7.1V and 17V be better (and why)?

Also, regarding the 5V USB wall wart what would be the optimum parameters for Current Output (Max/Min Amps)? Would a wall wart between 200 mA and 1000 mA suffice?

An externally hosted image should be here but it was not working when we last tested it.
 
There are 4 pins on the PCB for connecting the battery supply. +/-/+/- call them 1 to 4 so two in middle. Basically connect each output of DC stepnup to 9v as if it were a battery. The board handles the establishment of the GND for you.

Crystal clear now; thank you.

What would be a recommended voltage setting for both of the MT3608 DC-DC Step Ups... 9V?

Also, regarding the 5V USB wall warts what would be the optimum parameters for Current Output (Max/Min Amps)? Would a wall wart between 200 mA and 1000 mA suffice?

If one desired to hook up a 6.33mm (1/4”) jack (via a lead from the board) besides “switched” what other parameters must be followed for this amp design… Number of Positions/Contacts; Internal Switches; and Voltage Rating (e.g. 2 positions; 3 conductors/5 contacts; 500V)?

If one desired to hook up a 3 or 5mm diameter LED what are the Forward Current (If); Forward Voltage (Vf); and Operating Voltage (VDC) parameters to keep within? For example, the BOM shows a 10 mA; 2.1 Vf; 1.8 - 3.3 VDC… would a 20 mA 6VDC LED still work in the circuit?
 
Last edited:
Some light bulbs are beginning to turn on in my head in regards to the AC-DC power supply in my previous post above



So a Switching Split Power Supply is required, such as this one here which thus means that two (2) DC barrel panels jacks would be required. Correct?

Hey! Sorry about the delayed replied, I've been tied up yesterday. Also this morning, but should be able to give you some more information this afternoon. Big thanks to xrk971 for helping out.

Hey good find on that +/-12 500mA wall wart! That would solve your problem. assuming switching noise doesn't get into the audio. I've been searching for *years* for a +/- single wall wart. Must be something new on the market.

The one trick here is you would need to use two *isolated* DC barrel jacks on your panel, meaning they must be plastic with no contact between the panel and either of the barrel jack electrical contacts. Otherwise you will get a short through the panel. Those are harder to find but they exist. :) I probably have some here.

I have to run again, but will post more later. :)
 
Thank you good sir; very much appreciated. Await patiently.

FYI: I am planning to use an Extruded Aluminum Case with Plastic End Panels (Hammond 1455R2202BK) so as to avoid shorts and for an easier install.

Regarding that Switching Split Power Supply ± 12V 500mA Split Supply from adafruit here is the data sheet

I was thinking if this power supply could work I would purchase two (2) of the following plastic panel mount DC barrel jacks also from adafruit. DC Barrel Jack Datasheet

I am also planning to purchase this nicely machined knob as well from adafruit.
 
Thank you good sir; very much appreciated. Await patiently.

All of that looks good! Adafruit is a good place to go for DIY hobby parts.

The plastic end panels on the that Hammond case would solve the problem of needing isolated DC barrel jacks but... those DC jacks you have posted ARE the isolated type! :) The hard to find ones. Those would even work on a metal panel. Notice how all 3 electrical connectors come out the back of the those jacks so none touch the mounting panel, which is the key. The more common type has one of the connectors as a metal strip along the side of the jack, which would in turn contact with the mounting panel. Last I looked (a year ago) JDS also sells the isolated DC power jacks. I have them too.

In the DC power jack datasheet note that the current specification is 500mA and voltage 12V, which matches up perfectly with each half of that dual +/- power supply wall wart.

So really simple here, each one of the two pairs of wires coming out of that dual DC power adaptor replaces one of the battery snaps on the Super CMOY board. Just be careful with polarities. The Adafruit description tells you which part of each jack is + and which is -. I have + and - marked on the PC board, also you can just note the red (+) and black (-) wires in Alex' build.

The power adapter datasheet is leaving out one very important item, the switching frequency. I'm going to write Adafruit and see if that can get (or measure) that (most likely they can't/won't, so don't wait around, lol). The switching frequency is a clue as to if you will get audio noise from it. If the frequency is up in the 100's of Khz or a Mhz, probably no trouble. But if it is close to the audio range, like 40KHz, could be big trouble (switching noise inthe audio). Adafruit generally knows what they are doing, so I would expect this won't be a problem.

Also a comment about price. You might be thinking why are those packaged isolated converters from Mouser/Digikey $40 - $60 and this dual unit from Adafruit $15? For the wall wart they don't have to make it small, so they can use discrete parts on a (larger) PC board, which drops the price. They don't need a super-minature transformer, which would require the (high) 500KHz - 1 Mhz type switching frequency, to make the whole thing super small. Lol, I'm going to buy a few of those dual +/- power supply adapters to have on hand. :)

As for Super CMOY current requirements, each half of the OPA1688 is good for 75 mA, so 150 mA total, plus maybe 30mA queiscent in the chip and that LED. The dual chip version would be 2x, 300MA total. Either way well under that 500mA per rail from the dual power supply adapter.

Also interesting about that particular Hammond case, must be something new, it clearly is intended to be a knock-off of the Box Enclosure cases. :) Good to know, that might be an interesting alternative for projects. I've have to take a look.

In general those USB wall adaptors made for cell phones in your posts above wouldn't work. Neither the dual (two USB ports in one adaptor) or two singles. The trouble is lack of *isolation*, which means no electrical pathway between the two USB ports. The dual version would almost certainly have the - of the two connected together. Then there is a chance the single may have the - connected to the neutral of the incoming power.

In general to get isolation between the two power supplies you need transformer windings. Those isolated DC converters I've mentioned in the past have exactly that inside. Two separate, isolated, secondary windings with no DC electrical path between them. One powers the positive rail, the other the negative rail. This Adafruit dual supply adapter must also have a similar isolation transformer in there. These switching supplies run at high frequencies, like 100Khz - 1 Mhz, which drops the size of the transformer required.

The voltages and currents coming out of the Super CMOY are so low just about any 3.5mm or1/4" jack you find will meet specifications. As for the number of contacts, look at the symbols in my schematic for the Super CMOY and match them up with your jack. The input jack needs 2 contacts with two switches, so 5 connections total coming out of the jack (including ground). The output jack doesn't use the switches, so just the 3 connection type works (or the 5 with the switch contacts ignored, which is what the schematic shows).

As for the LED, you have small amount of flexibility there since I use a current source in that circuit. It will compensate for small variations in the LED forward voltage. So LEDs in the Vf=2V Vf=3.5V range (different colors, each color is a different VF) should work fine. LEDs have different rated sensitivities (the amount of light produced for a given current) - if the one you choose isn't bright enough, look for one that same color with with a higher sensitivity.

BUT... not 6V! Something is wrong there. LED forward voltages don't go up that high. The highest, as I recall, is white, which is around 3.9V or so. If your LED is rated for 6V it isn't just a LED, there is also a series resistor in the package so that the LED can be placed across 6V. That *won't* work here. Has to be just the LED in the circuit, no additional series resistance.

As for LED rated currents, on the LED datasheet you might see something like 10mA as the test current the forward voltage is quoted at but.. the LED can run at (way) less than 10mA and still produce light, which is the case in this circuit. It would be full brightness at 10mA, which may even be too bright to look at as a panel indicator, depending on LED sensitivity, but at say 2mA it still may produce a good amount of usable light. Just have to try them and see. :) But as rule of thumb, a LED with a test current of something like 40mA isn't going to work well (not produce usable light) at these smaller current. Stick with datasheet test currents in the 5ma - 20mA -or -so range on the datasheets.

Happy DIYing with your build!:)
 
Last edited:
All of that looks good! Adafruit is a good place to go for DIY hobby parts... ...Happy DIYing with your build!:)
Thank you for the detailed response regarding all my questions. Looking forward to the build.

The USB and DC-DC converter can only work if you have two of them - two ungrounded wall plug adapters. That will produce isolated 5vdc that can be put in series.
Thank you as well xrk971 for all the help and ideas.
 
Would someone kindly "double-check" the following desktop power supply wiring diagrams:

POWER SUPPLY PLAN A

An externally hosted image should be here but it was not working when we last tested it.


I am slightly confused about the white lead from the power supply... it being the inverse of the black lead means that I should wire the DC Barrel Jack to the PCB in reverse as well correct?


POWER SUPPLY PLAN B (if Switching Split Power Supply above is noisy)

An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
Hey....Thailand...thats quite a distance from here! lol...

Its on its way and I hope it arrives intact and ready to work for you!!

This sounds like an interesting project for sure.....its a really good amp. I forgot to mention it has the better tolerance WIMA caps in it.

All the best
Alex
 
Would someone kindly "double-check" the following desktop power supply wiring diagrams:

I'm on the phone internet and will have a better look later, but your plan A looks just fine. Plan B...that doesn't look like a dual isolated converter from the pinout diagram, just a single output isolated. There should be two independent pairs of outputs in the converter that you can wire like you did in plan A.
 
I'm on the phone internet and will have a better look later, but your plan A looks just fine. Plan B...that doesn't look like a dual isolated converter from the pinout diagram, just a single output isolated. There should be two independent pairs of outputs in the converter that you can wire like you did in plan A.
Ahhhh, ok I thought something was looking fishy there...

I combed the internet but I am unable to find a pin out diagram for the RW-0509D (DIP-24 Miniature) DC-DC converter but I think it will be something like this below.

Four (4) resistors (499D) would be required to be soldered between each output (11, 12, 13, & 14) and their respective common ground (9 or 16).

POWER SUPPLY PLAN B

An externally hosted image should be here but it was not working when we last tested it.



After reading up on potential/possible frequency noise issues associated with Switching Power Supplies I am now finding myself leaning to Plan B. Probably the safer and surer bet.
 
Last edited: