but dont lion batteries need some very specific charge circuitry as well to stop them...ermm....exploding?
that was more my point, my comments werent really directed at the se-se version as i use a portable amp with basically that same combination plus stacked buffers every day, but if you want to use high capacity batteries and fast charge them, there is more involved than just putting some batteries in a case unless you want to take them out to charge each time.
hmmm portable balanced buffalo 2 with opa1632, transformer coupled toslink and mini bnc you say........
2300mah +/-6v6 (13.4v total) lifepo4 battery you say?
driving jh13 you say?
sounds insane, noone in their right mind would carry that around......
i'm actually interested to see if i can squeeze the bal-bal version in and still get reasonable battery life. although i wouldnt have to populate the opa1632, as thats whats in there already (dgn version), but no buffer, either that or i'm looking at perhaps making some little daughterboards to put a buffer in the loop. low z drive is already very good, but you know how it is
i use an external balance charger from thunderpower, used for radio control battery packs, charges each cell separately which is needed for me as most of the draw is on the positive side, with only about 120ma out of 600ma being negative
i use an old iriver ihp120 modded for coax spdif out and with a 32gb compact flash card instead of the hd
that was more my point, my comments werent really directed at the se-se version as i use a portable amp with basically that same combination plus stacked buffers every day, but if you want to use high capacity batteries and fast charge them, there is more involved than just putting some batteries in a case unless you want to take them out to charge each time.
hmmm portable balanced buffalo 2 with opa1632, transformer coupled toslink and mini bnc you say........
2300mah +/-6v6 (13.4v total) lifepo4 battery you say?
driving jh13 you say?
sounds insane, noone in their right mind would carry that around......
i'm actually interested to see if i can squeeze the bal-bal version in and still get reasonable battery life. although i wouldnt have to populate the opa1632, as thats whats in there already (dgn version), but no buffer, either that or i'm looking at perhaps making some little daughterboards to put a buffer in the loop. low z drive is already very good, but you know how it is
i use an external balance charger from thunderpower, used for radio control battery packs, charges each cell separately which is needed for me as most of the draw is on the positive side, with only about 120ma out of 600ma being negative
i use an old iriver ihp120 modded for coax spdif out and with a 32gb compact flash card instead of the hd
Last edited:
I could put in some information about the lithium ion/polymer batteries. Used some for my masters thesis a few months ago and did some study on it to make sure that it didn't blow up in my face...
First the ion/polymer chairs the same major characteristics the main difference is the package. Mostly the Li-ion are round cells found in laptops and the Li-pol are the flat types found in cellphones and other similar devices.
Li-ion/pol batteries are actually really easy to maintain compared to the Ni- based types since the nickel based one require a very specific charge/discharge cycle for optimal performance.
Simple rules about the Li-ion/pol batteries:
1. NEVER charge each cell above 4.2 V, see (*)
2. Never charge the cell in temperatures below 0 degrees Celsius, lithium oxide will be plated on the anode and next time you charge (*) can/will happen.
3. Discharge in low temperature is not an issue except for a decrease in capacity but the cell won't be damaged you just need to charge faster.
4. Never let the cell go below 3 V (check datasheet). This will permanently damage the cell and in the best case you lose capacity, worst case leads to (*) next time its charged...
5. If you connect several celles in series some form of balancing circuit is required to make sure no cell goes over 4.2 V
6. Tip, if the batteries are only charged to 4.1 V the lifetime will be prolonged and its nice to be on the safe side
7. Don't charge them faster than about 0.7 C (where C is the capacity of the battery).
8. If a cell is swollen, that means that oxygen has been produced and then you shouldn't use it anymore. It might work at decreased capacity but its just waiting for a reason to blow.
9. Don't draw to much current from the cell since that could cause (*) in the extreme case. I think the limit is usually around 0.5-1 C which should be safe for this application unless you get a short...
10. The positive part, some cells have an internal protection circuit which helps but not all so check...
(*). The cell will start to produce oxygen gas which in turn expands the pack until it ruptures and then you have a perfect mix of oxygen, highly reactive lithium and quite a lot of electrical energy which means sparks to trigger the reaction even faster.
To charge a single cell first limit the current to a maximum of 0.7 C until the voltage reaches 4.2 V (-0.1 V for safety). The keep the voltage at maximum until the current drops to ~10 mA (if i remember correctly). Then you are done. There are some neat chargers for these type of cells to you don't have to do everything from scratch. I used LTC3105, really nice for single cell applications using solar cells for charging (won't work well in this case).
There are a few rules to think about but a voltage and current limitation is simple compared to what the nickel batteries require. And the idle lifetime for a lithium based cell is a lot better than the nickel based ones.
First the ion/polymer chairs the same major characteristics the main difference is the package. Mostly the Li-ion are round cells found in laptops and the Li-pol are the flat types found in cellphones and other similar devices.
Li-ion/pol batteries are actually really easy to maintain compared to the Ni- based types since the nickel based one require a very specific charge/discharge cycle for optimal performance.
Simple rules about the Li-ion/pol batteries:
1. NEVER charge each cell above 4.2 V, see (*)
2. Never charge the cell in temperatures below 0 degrees Celsius, lithium oxide will be plated on the anode and next time you charge (*) can/will happen.
3. Discharge in low temperature is not an issue except for a decrease in capacity but the cell won't be damaged you just need to charge faster.
4. Never let the cell go below 3 V (check datasheet). This will permanently damage the cell and in the best case you lose capacity, worst case leads to (*) next time its charged...
5. If you connect several celles in series some form of balancing circuit is required to make sure no cell goes over 4.2 V
6. Tip, if the batteries are only charged to 4.1 V the lifetime will be prolonged and its nice to be on the safe side
7. Don't charge them faster than about 0.7 C (where C is the capacity of the battery).
8. If a cell is swollen, that means that oxygen has been produced and then you shouldn't use it anymore. It might work at decreased capacity but its just waiting for a reason to blow.
9. Don't draw to much current from the cell since that could cause (*) in the extreme case. I think the limit is usually around 0.5-1 C which should be safe for this application unless you get a short...
10. The positive part, some cells have an internal protection circuit which helps but not all so check...
(*). The cell will start to produce oxygen gas which in turn expands the pack until it ruptures and then you have a perfect mix of oxygen, highly reactive lithium and quite a lot of electrical energy which means sparks to trigger the reaction even faster.
To charge a single cell first limit the current to a maximum of 0.7 C until the voltage reaches 4.2 V (-0.1 V for safety). The keep the voltage at maximum until the current drops to ~10 mA (if i remember correctly). Then you are done. There are some neat chargers for these type of cells to you don't have to do everything from scratch. I used LTC3105, really nice for single cell applications using solar cells for charging (won't work well in this case).
There are a few rules to think about but a voltage and current limitation is simple compared to what the nickel batteries require. And the idle lifetime for a lithium based cell is a lot better than the nickel based ones.
yep, exactly why i chose lifepo4, overcharge will not cause significant damage, can even sometimes trickle charge. can fast charge in a matter of minutes (up to 40c i believe), lasts considerably longer, will put out 40 amps constant, with burst of 130amps, lower output impedance, lower noise. shall i go on?
basically the main disadvantage is they dont have quite the energy density of lipoly, but thats it. still far higher density than anything else around
and yes, if going for battery, i realy once again HIGHLY recommend the linear tech stuff. basically my unit has 12 onboard regulators and all operating current is controlled in dropout, the bipolar regs i'm using (lt3032) operate on less than 1ma. these are the guys that are pioneering the field of RF energy harvesting. battery powered safety gear and locator devices for building sites and mines that operates on nothing but the radio signals in the air, just plucks it right out of the air
they also create what i feel are the best, most complete application notes and datasheets in the business
basically the main disadvantage is they dont have quite the energy density of lipoly, but thats it. still far higher density than anything else around
and yes, if going for battery, i realy once again HIGHLY recommend the linear tech stuff. basically my unit has 12 onboard regulators and all operating current is controlled in dropout, the bipolar regs i'm using (lt3032) operate on less than 1ma. these are the guys that are pioneering the field of RF energy harvesting. battery powered safety gear and locator devices for building sites and mines that operates on nothing but the radio signals in the air, just plucks it right out of the air
they also create what i feel are the best, most complete application notes and datasheets in the business
Last edited:
i wonder how the portable version of this amp would look..
judging from the pcb and maximum height of the components, i don't think this amp can make it in the case below 3cm. higher than that, i'd rather call it 'transportable' since the dimension will not be good enough to be carried in your hand
judging from the pcb and maximum height of the components, i don't think this amp can make it in the case below 3cm. higher than that, i'd rather call it 'transportable' since the dimension will not be good enough to be carried in your hand
Last edited:
If you use the Nichicon low profile 8mm polymer caps, the total PCB height including components would be less than 10mm. If you get crafty and solder some SMD caps on the PTH pads, you could probably get that even lower. You'd have to use 3.5mm jacks for input and output to keep things small, but that's probably what you'd want for a portable setup anyway.
Batteries are a whole other can of worms, but there are cell phone batteries out there that are less than 5mm thick, so you could potentially make something that is very low profile.
Cheers,
Owen
Batteries are a whole other can of worms, but there are cell phone batteries out there that are less than 5mm thick, so you could potentially make something that is very low profile.
Cheers,
Owen
At the university we have a nice oven or a soldering iron with hot air for those things. The hot air works but you will cook the poor chip. This is for the people with fancy equipment, there are a few cool DIY tricks which could be possible. I haven't tried these myself since i haven't had the need to.
1. Fill a frying pan with dry finegrade sand and get a thermometer to keep the temperature under control. Apply paste to the pads and place the component in the right place and gently place the PCB in the sand when the temperature is high enough. Paste is best but pretinned pads and a lot of flux might work as well.
2. I heard some people using a normal toaster, they got results (decent?) but the frying pan won that competition easily if i remember correctly.
3. Make the layout as usually but make sure you have no pads under the component. Drill from the bottom to the top copper layer so you can heat the pad using a normal soldering iron from the bottom of the PWB.
Personally i would make the ground plane out from the sides so its enough to use a wide tip from the sides to heat the entire ground plane. This is not good since it will require a lot of energy and can damage both the board and component but its an simple solution to test and it could work.
The signal pads are actually quite easy to solder using a lot of flux and a small amount of solder on the tip and just push it against the pad. The DFN packages are nice in this way because you can see the pads from the side they are not completely hidden underneath. Still its quite tricky but possible if you are careful.
If you make your own layout make the pads a bit longer, this will help if you want to solder manually.
Wow. Cool, but a lot of work.
I took the lead from L-Train:
http://www.diyaudio.com/forums/head...-headphone-amplifier-pcbs-26.html#post2478234
Worked just fine. Just be patient. Once the flow starts, it works like a charm.
I took the lead from L-Train:
http://www.diyaudio.com/forums/head...-headphone-amplifier-pcbs-26.html#post2478234
Worked just fine. Just be patient. Once the flow starts, it works like a charm.
Here is something fun. I just calipered an Altoids can at 3.55" L x 2.50" W x 0.80" D. Those are to the rolled edges, slightly more inside the can itself.
These 9V 1200mAh (1.2Ah) lithium batteries are 1.92" L (add a bit for the connector) x 1.02" W x 0.67" D - they are a "caseless" (no aluminum shell) version:
U9VL / U9VL-FP - NSN 6135-01-369-9792 : 9-Volt Batteries - Ultralife Battery & Energy Products | Ultralife Corporation
Ultralife 9V Lithium Battery U9VL-FP U9VLFP6 Foil Pack, Long Life, Plastic Casing Foil Pack
Two side by side in the Altoids can (one per rail, of course) would be 2.02" or thereabouts. The SE->SE board sideways is 1.25", giving a total width of the PCB and two batteries at 3.27". It all fits!
That is assuming the 3.5mm jacks can fit in the 0.40" above or below the PCB. I'm assuming the batteries would be removed for external charging.These cells have a maximum discharge rate of 120 mA. The discharge graph from the "charts" tab shows a fairly steady 8.0V after the initial drop off from 10V out to 0.8Ah capacity assuming a steady load of 90R. 8V / 90R = 90mA draw (per rail). 0.8 Ah then would be 0.8Ah / 0.09A = 8.9 hour (per battery / rail).
8V steady state per rail minus the (Vcc-2.1V) overhead from the LME49600 spec sheet at 100mA gives 5.9V peak, or 4.1V rms. Into my 44 ohm Shure SRH940 headphones that gives 4.1V / 44R = 93mA, which almost matches up with the 90mA discharge assumption, except then add the quiescent for both chips of 20mA (low BW mode on the buffer) and you are up to 110mA.
That is still under the 120mA max discharge limit for the cell but would require another iteration through the math. Likely resulting in a drop in steady state voltage to around 7.8V and reduction in time to 6 hours.
But that is with nearly absolute maximum discharge per cell. 4.1V to the Shures would be... [(4.1V)^2] / 44 = 380 mW, lol! The Shures are actually rated for that, 1000mW total, but I'm certain my ears are not. So the amp would not be run anywhere near that output voltage / volume level. The real world max discharge rate would likely be much less than 90mA to the phones (per rail), maybe half that at 50mA, meaning a much longer battery life. Making a 70% "music power" assumption would drop the average discharge rate even lower and makes the battery life even longer.
Better than a cmoy.... it is an opc!
Last edited:
by designing the pcb yourself to cater for this problem. this reg was so very appealing that i thought long and hard about the problem. it is solved by creating a custom pattern for the part. the pads for power pads underneath have pth vias that link to the correct plane/copper pour (its a 4 layer board) and after tacking/gluing in place, very lightly tinning the pad on the device and adding a bit of flux, solder from the bottom of the board. the side 'leads' also extend up the side face a touch, so the pattern has extended the leads out a further 2mm, flux, solder. soldering a 14 pin part that is only 3 x 4mm is another issue
but doable.of course you can (and i will) build toaster over reflow controllers for not much money, i do enough smd to make this an interesting idea.
yeah owen, thats what i did with the buffalo 2 as well, the oscons have been replaced with panasonic SP caps and tantalum/polymer 7343 package. thats the package all the high performance lowz caps are in these days it seems. cheaper than sanyo oscon, but not high enough voltage for this board, maybe the input stage, there are ones that are 20v+ but not the really high grade ones
@ thegreatbrian
whatever suits you and your gear, its not actually designed with an input pot in mind (at least v1 wasnt), but you will apply one in the voltage gain (input) opamp feedback loop. i presume this time opc has put some pads for this?
regardless a good place to start is 25k, this is not too high for noise, but also not too low so should be well driven by just about anything.
whatever suits you and your gear, its not actually designed with an input pot in mind (at least v1 wasnt), but you will apply one in the voltage gain (input) opamp feedback loop. i presume this time opc has put some pads for this?
regardless a good place to start is 25k, this is not too high for noise, but also not too low so should be well driven by just about anything.
Wouldn't I be able to just put the attenuator between the inputs and connect it to the input pads of the pcb like in post #440?
you could, but personally i would install in the loop. either will work. actually you would probably want to go lower than 25k if you put it in the loop and you would want to adjust the other gain resistors to suit. i feel its the better way to do it, but i suppose its a matter of taste
Last edited:
noise created by the resistors and contacts is to a degree averaged out with feedback, no extra resistors are added, as they simply replace those that are already there.
one problem is that you must find a way to achieve having the pot mechanism as close to the opamp as possible, extending the feedback loop to the front panel and back is not such a great idea.
best way imo would be a relay based attenuator in the loop, this allows very short paths.
again, all a matter of taste
one problem is that you must find a way to achieve having the pot mechanism as close to the opamp as possible, extending the feedback loop to the front panel and back is not such a great idea.
best way imo would be a relay based attenuator in the loop, this allows very short paths.
again, all a matter of taste