IMPORTANT! As reported by Schmidtc89 and forsakenrider, the front right corner of the gain switch, S2, can possibly touch the via (hole in the PCB) located next to R21. This will cause problems with one channel of the amp but won't cause any harm. It would seem most boards are not having the problem but it's worth taking some precautions.
I should have caught this, but it's the sort of risk one runs in ordering a large batch of PCBs before hardly anyone has built amps. The via used to be under the "foot" towards the middle of the switch so I moved it up close to the edge of R21 thinking it was out of the way (it looks fine in the overlay view of the board in software). But looking at my assembled board, I can see it's too close to the corner "foot" of S1. It's not shorting on my my board but I could see how it could. At least the short won't harm any components, put DC on the output, or do anything else dangerous.
The solder mask registration on boards tends to vary a bit from board to board. The solder mask on a given board, how the switch is installed with the bit of play present in the holes, and if it's fully seated down on the board, will determine if it could short.
THE FIX FOR ASSEMBLED BOARDS: If you have a board with a shorting gain switch, the easiest way to fix it is to use a small flat screwdriver or similar under that corner of the switch and heat the pin, or pins, in that corner of the switch (closest to R21) while prying up on the switch. It should move enough to lift the "foot" off the via. You can easily test with a DMM or just pry it up until you can see a sliver of "daylight" under the foot.
FOR THOSE WHO HAVEN'T YET BUILT THEIR BOARD: If you have some fairly beefy diagonal cutting pliars you can trim the leg on that corner of the switch. Don't try to use "precision" cutters as you may damage them. The other option is to bend the foot inward slightly or simply install the switch with that corner raised up slightly.
I'll put a note right now in the assembly instructions and, ideally, I should also revise the artwork to move the via but I'll wait for some more feedback from O2 builds to see if anyone discovers anything else that needs tweaking. Sorry for any hassles and thanks for bringing this to my attention. Please feel free to pass this along in other forum threads, etc. Thanks.
I should have caught this, but it's the sort of risk one runs in ordering a large batch of PCBs before hardly anyone has built amps. The via used to be under the "foot" towards the middle of the switch so I moved it up close to the edge of R21 thinking it was out of the way (it looks fine in the overlay view of the board in software). But looking at my assembled board, I can see it's too close to the corner "foot" of S1. It's not shorting on my my board but I could see how it could. At least the short won't harm any components, put DC on the output, or do anything else dangerous.
The solder mask registration on boards tends to vary a bit from board to board. The solder mask on a given board, how the switch is installed with the bit of play present in the holes, and if it's fully seated down on the board, will determine if it could short.
THE FIX FOR ASSEMBLED BOARDS: If you have a board with a shorting gain switch, the easiest way to fix it is to use a small flat screwdriver or similar under that corner of the switch and heat the pin, or pins, in that corner of the switch (closest to R21) while prying up on the switch. It should move enough to lift the "foot" off the via. You can easily test with a DMM or just pry it up until you can see a sliver of "daylight" under the foot.
FOR THOSE WHO HAVEN'T YET BUILT THEIR BOARD: If you have some fairly beefy diagonal cutting pliars you can trim the leg on that corner of the switch. Don't try to use "precision" cutters as you may damage them. The other option is to bend the foot inward slightly or simply install the switch with that corner raised up slightly.
I'll put a note right now in the assembly instructions and, ideally, I should also revise the artwork to move the via but I'll wait for some more feedback from O2 builds to see if anyone discovers anything else that needs tweaking. Sorry for any hassles and thanks for bringing this to my attention. Please feel free to pass this along in other forum threads, etc. Thanks.
Last edited:
If your goal is only charging, you can save some money and hassle and use a single 24 volt DC-DC converter with a series diode and resistor connected directly across both batteries. The resistor should be in the range of 270 - 390 ohms depending on if you plan to leave the supply connected for extended periods. The diode should be a 1N5818. That way you could use a panel mount regular coaxial DC power jack for the charging input. The downside to this approach is the external supply (using the back panel connector) will not run the amp and charge the batteries at the same time like the AC power supply will. This will also work with any 24 VDC wall transformer.
I'm not sure what to suggest for a small inexpensive 3 pin panel mount connector that won't short out the DC-DC converter (like a 3.5mm plug would). There might be room for a Mini DIN jack but I'm not sure. The other small multipin connectors tend to get expensive, require crimped pins, etc.
Thanks, everyone, for the heads-up on S2. I planned a mod that make it a virtual non-issue, but it's good to know.
The 24V recharge is something I hadn't thought of. It's an option to be considered. I'd like to include a fused USB charge option, so I'm not sure how workable that is from 5V. Like the OPA2277, I already have the Murata converters. Mini-DIN or a FILO 3.5mm jack were the best I had come up with. I don't want a makeshift solution in case there's enough interest to post it for the DiyAudio gang.
The 24V recharge is something I hadn't thought of. It's an option to be considered. I'd like to include a fused USB charge option, so I'm not sure how workable that is from 5V. Like the OPA2277, I already have the Murata converters. Mini-DIN or a FILO 3.5mm jack were the best I had come up with. I don't want a makeshift solution in case there's enough interest to post it for the DiyAudio gang.
I have one like this My Soldering Iron that I usually use around the 1:30 - 2:00 position, so around 700 degrees? Although I have no idea if it's a linear adjustment..
I'll have to see if the IR temp sensor I just ordered can tell me what the tip temp is..
I'll have to see if the IR temp sensor I just ordered can tell me what the tip temp is..
It's trial and error as few irons have a very accurate calibrated scale (assuming they don't have a digital temp readout). Around 650 is hot enough for most connections and helps keep the tip from turning black just sitting in the holder. 700 is better for really big connections (like the battery connectors on the O2) where you need lots of heat. But the tip will oxidize faster if you run at 700 all the time. Anything much over 700 will really shorten the tip life and is too hot for delicate connections.
It also depends if your iron has true closed loop feedback and how fast it corrects the tip temp. 650 is plenty on a pro-grade iron as it will maintain that temp even when soldering very large connections. But cheaper irons will "lag" and it will take much longer to solder big connections at 650.
If you're going to take break you can turn the iron down to the minimum setting (rather than turning it off) for faster warm up when you come back. Some pro irons do that automatically. They can tell when you haven't soldered anything in a while and they go into standby.
All things considered, I still use a soldering "gun". One of those gun-shaped soldering irons, with no temp control but the trigger. And I solder anything from SMD to really large joints with it.
What I'm saying is, don't be *too* fussy about it. If you notice you're turning black and oxidizing too fast, lower it a bit. If it doesn't melt solder *instantly*, raise it up a bit. After a few tries you'll know how to tune it for what you want.
Anyone wants to buy me a closed-loop feedback soldering station for Chrstmas?
What I'm saying is, don't be *too* fussy about it. If you notice you're turning black and oxidizing too fast, lower it a bit. If it doesn't melt solder *instantly*, raise it up a bit. After a few tries you'll know how to tune it for what you want.
Anyone wants to buy me a closed-loop feedback soldering station for Chrstmas?
I used a cheapo radio shack soldering iron and didn't have any major problems. The only time I had difficulty was when I as trying to solder the MOSFET, the MOSFET heat-sinks would pull the heat away almost too fast, so getting the solder to stick to the board and pins was "fun".
Methinks something may be wrong in my power management circuit.
When performing the "Low Voltage Shutdown" test, here are the numbers I get:
Battery #1 connected, no AC power, amp on
233mV across P2 (Square pad to both #3 & #4 pads)
479mV across U3 pins 4 & 8
Battery #2 connected, no AC power, amp on
323 mV across P2 (square pad to #3 pad)
421 mV across P2 (square pad to #4 pad)
43mV across U3 pins 4 &8
I used the same 9V battery for both positions, just to make sure the difference in numbers wasn't due to different batteries.
What should be my first course of action? Replace Q1 and/or Q2?
Thanks for your help!
When performing the "Low Voltage Shutdown" test, here are the numbers I get:
Battery #1 connected, no AC power, amp on
233mV across P2 (Square pad to both #3 & #4 pads)
479mV across U3 pins 4 & 8
Battery #2 connected, no AC power, amp on
323 mV across P2 (square pad to #3 pad)
421 mV across P2 (square pad to #4 pad)
43mV across U3 pins 4 &8
I used the same 9V battery for both positions, just to make sure the difference in numbers wasn't due to different batteries.
What should be my first course of action? Replace Q1 and/or Q2?
Thanks for your help!
Hi there, I'm still having trouble fixing some problems with my O2 build.
I too am having problems with the power management part of my amp, except mine is a bit more messed up.
When I try the low voltage shutdown test, and connect one 9V battery at BT1, and measure pin 1 and 4 of P2, it shows 229mV, but when I measure U3 pin 4 and pin 8, I get 470mV. Conversely when I do the same test with BT2 connected only, the voltage builds up slowly from 20mV to 40mV when measuring P2, and while the U3 test measures 56.5mV.
Also, when I use the AC to power the amplifier, I can't use the power switch to turn it on, I have to unplug and replug the AC plug to make it turn on, and after turning it on, U5 and U6 heats up quite a bit, to say, the temperature of a cup of coffee. no idea if that's normal or not.
I suspect one or both of my MOSFETs are damaged, but I have no idea if it's true (newbie here), and/or how to test if they're working or not. Thanks for any help.
I too am having problems with the power management part of my amp, except mine is a bit more messed up.
When I try the low voltage shutdown test, and connect one 9V battery at BT1, and measure pin 1 and 4 of P2, it shows 229mV, but when I measure U3 pin 4 and pin 8, I get 470mV. Conversely when I do the same test with BT2 connected only, the voltage builds up slowly from 20mV to 40mV when measuring P2, and while the U3 test measures 56.5mV.
Also, when I use the AC to power the amplifier, I can't use the power switch to turn it on, I have to unplug and replug the AC plug to make it turn on, and after turning it on, U5 and U6 heats up quite a bit, to say, the temperature of a cup of coffee. no idea if that's normal or not.
I suspect one or both of my MOSFETs are damaged, but I have no idea if it's true (newbie here), and/or how to test if they're working or not. Thanks for any help.
I just got the invoice from Newark with shipping etc included: shipping was $12 and tax (for 2 B2 enclosures) was $5.42.
? Substitute parts ?
I have recently got my order of parts from Mouser - unfortunately, one critical part - the resistor, part #34 (2.74 Meg 1/8 Watt <= 5% Film 1.9mm < 3.9mm) - is on backorder and may not be available for a couple of months. Digikey have the part, but have a £12 postage charge on small orders.
What is the acceptable range of values / tolerances that could be used as a substitute for this part - ideally so that I can source it locally.
I have recently got my order of parts from Mouser - unfortunately, one critical part - the resistor, part #34 (2.74 Meg 1/8 Watt <= 5% Film 1.9mm < 3.9mm) - is on backorder and may not be available for a couple of months. Digikey have the part, but have a £12 postage charge on small orders.
What is the acceptable range of values / tolerances that could be used as a substitute for this part - ideally so that I can source it locally.
@Eric & Tschuss, it sounds like everything is OK. I revised the testing instructions a week or so ago to point out you may see several hundred millivolts at the output with no load in shut down (or even with the amp turned off with the power switch). To confirm your amps are OK, you can try this:
1 - Make sure there is less than 600 mV across C8 and C9 when the amp is trying to run from one battery.
2 - Connect some "junk" headphones you don't care too much about (they should be fine) to the output jack, wait a few seconds, then measure at P2 again. This puts a load on the amp and helps drain off the little bit of charge remaining in C8 and C9.
Generally the MOSFETs in the power management circuit are either fully on or fully off. So you'll get 9+ volts at C8 and/or C9 or you'll get less than 0.6 volts. The reason you're seeing several hundred millivolts is the transistors in the all the op amps stop working below 500 mV. So there's little to drain the remaining charge from C8 and C9 if there's no load on the amplifier. Most amps with direct coupled outputs do the same thing. With real world headphones connected the slight offset goes away very quickly.
I'll look at further revising the test procedure to make it more clear.
1 - Make sure there is less than 600 mV across C8 and C9 when the amp is trying to run from one battery.
2 - Connect some "junk" headphones you don't care too much about (they should be fine) to the output jack, wait a few seconds, then measure at P2 again. This puts a load on the amp and helps drain off the little bit of charge remaining in C8 and C9.
Generally the MOSFETs in the power management circuit are either fully on or fully off. So you'll get 9+ volts at C8 and/or C9 or you'll get less than 0.6 volts. The reason you're seeing several hundred millivolts is the transistors in the all the op amps stop working below 500 mV. So there's little to drain the remaining charge from C8 and C9 if there's no load on the amplifier. Most amps with direct coupled outputs do the same thing. With real world headphones connected the slight offset goes away very quickly.
I'll look at further revising the test procedure to make it more clear.
@RocketScientist-
Thank you for responding. I'm doing the test with the power switch of the amp turned on. I think that's the way to do it according to your instructions. So when you say "shut down", you're actually talking about the power management circuit doing its job....NOT that the amp is actually switched off, correct?
Also, should I be concerned that the two battery positions give such wildly different readings? That was actually my primary concern. I couldn't figure out why there was a ~10x difference across U3 pins 4 & 8 depending on whether B1 or B2 was populated.
Thanks again for following up. It's awesome to have this kind of support for what is essentially a "free" product.
Thank you for responding. I'm doing the test with the power switch of the amp turned on. I think that's the way to do it according to your instructions. So when you say "shut down", you're actually talking about the power management circuit doing its job....NOT that the amp is actually switched off, correct?
Also, should I be concerned that the two battery positions give such wildly different readings? That was actually my primary concern. I couldn't figure out why there was a ~10x difference across U3 pins 4 & 8 depending on whether B1 or B2 was populated.
Thanks again for following up. It's awesome to have this kind of support for what is essentially a "free" product.
@tschuss, I wouldn't worry about the differences as long as they're all under 0.6 volts. As long as the offset at P2 (or the headphone jack) falls below 20 mV with headphones connected everything is fine.
Yes, by "shut down" I mean the power management circuit shutting down the audio portion of the amp. But, with no headphones connected, you'll see similar voltages if you simply turn the amp off with the power switch which completely disconnected the batteries and AC power from the amp. The voltages are just a tiny amount of residual charge in C8 and C9 that lingers after the op amps all stop operating.
Yes, by "shut down" I mean the power management circuit shutting down the audio portion of the amp. But, with no headphones connected, you'll see similar voltages if you simply turn the amp off with the power switch which completely disconnected the batteries and AC power from the amp. The voltages are just a tiny amount of residual charge in C8 and C9 that lingers after the op amps all stop operating.
@RocketScientist, thanks for your reply as well. I guess that's one less problem for me to worry. However, do you have any clue what's going on with my AC power issue? When I connect the AC adapter only, and attempt to turn on the amplifier through the switch S1, it doesn't power on, despite pressing the switch numerously, however after unplugging and replugging the power plug with the power switch at the "on" position, the amplifier turns on successfully automatically and works perfectly, albeit U5 and U6 being hot (which i have no idea if that's normal).
I tried your steps that you gave above, and it's all inside normal parameters, but does that mean both MOSFETs are working, and isn't related to my issue?
I tried your steps that you gave above, and it's all inside normal parameters, but does that mean both MOSFETs are working, and isn't related to my issue?
ericlaw02 - a couple of things to try while you are waiting for RocketScientist's next login...
Do you have a DMM to measure voltages with? If so what is the voltage reading from ground (metal body of the gain switch) to the positive rail (highest voltage side of R1), and then from ground to the negative rail (lowest voltage side of R2), with the switch S1 off and no batteries installed (ac adaptor plugged in of course)? Then what are the same two readings with switch S1 on?
Are U5 and U6 so hot that you would get burned to keep a finger on them, or not quite that hot that you could leave a finger on them for 10 sec or so?
During one of the times when the adaptor is plugged in, switch S1 is on, but the amp is off try to measure the voltage between the negative rail (lowest side of R2) and pin 1 of U2. Then between the negative rail and pin 7 of U2. The U2 pins are numbered counterclockwise looking at it from the top, with 1 in the upper left corner by the dot. Be very careful not to short IC pins during testing!
Do you have a DMM to measure voltages with? If so what is the voltage reading from ground (metal body of the gain switch) to the positive rail (highest voltage side of R1), and then from ground to the negative rail (lowest voltage side of R2), with the switch S1 off and no batteries installed (ac adaptor plugged in of course)? Then what are the same two readings with switch S1 on?
Are U5 and U6 so hot that you would get burned to keep a finger on them, or not quite that hot that you could leave a finger on them for 10 sec or so?
During one of the times when the adaptor is plugged in, switch S1 is on, but the amp is off try to measure the voltage between the negative rail (lowest side of R2) and pin 1 of U2. Then between the negative rail and pin 7 of U2. The U2 pins are numbered counterclockwise looking at it from the top, with 1 in the upper left corner by the dot. Be very careful not to short IC pins during testing!
Last edited: