I would think the battery pack mentioned earlier would come with a board something like this. No, probably does not do the full charging/voltage regulation. Just giving it as an example of the size and functions available. I am guessing the people that put together the pack spent more than the 5 minutes I did looking for a board. 4S 30A Li-ion Lithium 18650 Battery Cell BMS PCB Protection Balance Board 14.8V 973777478191 | eBay
Last edited:
Thanks for that! It looks like MOSFET "on" resistance has now dropped so low that you can put a 30A pass element in SMD format on a tiny PCB with no heatsink, and no guarantee of forced air cooling. Impressive.
(At 30 amps, power dissipation is about one watt even with only one single milliohm of on resistance.)
I've seen 30A pass capability in a PCB the size of a postage stamp before, but only in applications that provided some forced air cooling.
-Gnobuddy
Well, I'm still after my original goal----the best emulation of a Princeton Reverb that can be battery-powered. That means 12 to 15 watts driving a 10 " Jensen (or Weber) 8Ω speaker. Now I don't know....12 volts DC powering a traditional Class A-B amp could only produce about 2 watts rms. But don't those Class-D amplifiers bump up the voltage to provide more power? Thay must do that in car amps, eh? I know their boasts of 200 watts and such are bogus, but surely they can at least do 12 watts, can't they? And they are operating from 12 volts. What is the best eBay special Class-D amp that will do an honest 12 watts into 8Ω from a 12 volt supply???
I think you need about 18V from your battery to achieve that, using any of the class-D boards that are commonly available.
Agreed (my estimate is actually even lower, around 1.25 W.)
However, all the recent class-D amp boards I've seen are actually 4 separate power amps in one chip - two pairs of amps, each pair wired in bridge mode. The speaker isn't grounded at either end, but is wired between the outputs of two amps, which are driven in opposite phases by the internal driver circuitry.
So now you have twice the peak voltage that you'd get without the bridge mode operation. Twice the voltage, ergo four times the power. The roughly 2 watts you mentioned becomes roughly 8 watts.
My estimates are a bit more conservative (I allowed for 3 volts lost in the output devices and battery voltage droop under load which gives you only 1.25 W on 12V supply.) But you also would have about 13 V (not 12 V) from that particular 4S LiFePO4 pack. So my estimate was about 6 watts RMS.
Without bridge mode, you can only get some 1.25 clean watts out of a 12 V supply and 8 ohm speaker.
Going bridge mode bumps that to about 5 watts RMS. Dropping to a 4 ohm speaker gets you around 10 watts RMS. There are typically 4 speakers roughly at the 4 corners of the car interior, so now you have around 40 watts RMS, in a "room" that is really very small.
I think a lot of stock car audio systems just leave it at that, or maybe use custom speakers with even less than 4 ohms impedance. For typical casual in-car listening, 40 watts in such a small space is more than loud enough.
If you want more than an honest 40 watts RMS in a car, then you have to use some sort of DC-DC boost converter to step up the voltage (or use weird low-impedance factory speakers.) I've seen aftermarket car stereos that have circuitry like this built in. But I haven't seen any DC-DC converter like this built into any of the usual cheap class-D power amp boards we see on Ebay, Amazon, and so on.
I don't know of one, maybe someone else does?
If you're willing to use 2 of those 10", 8 ohm Jensens, one driven by each of the 2 channels on the class-D board, then you are at 12 watts RMS - roughly six watts into each speaker - with a 13 volt supply (4 LiFePO4 cells in series.) But a 2x10 cab isn't tiny, or light!
Otherwise, if only 1 speaker is used, you need that 18 volt battery...
-Gnobuddy
Well, I found this:
DC 12V-24V TDA7492MV Mono 50W Class D Audio Power Amplifier Board Amp Module 699971085729 | eBay
and looked up the datasheet for the TDA7492MV chip from ST. The 50 watts claimed for the eBay board is ********, of course, but the datasheet indicates a fairly clean (1% THD) 12 watts into 8Ω with a 15 volt supply. That'll do me. Then my buddy found these:
4X 18650 3.7V 9800mAh TR Li-ion Rechargeable GTF Battery + Smart US Charger OY | eBay
8 of them in series-parallel would give me ~ 15 volts at 2 A-H and only cost about $16. Hmmmmm......
I'll end up spending more on the speaker than all the rest, but that is only right, methinks, because it makes the most diffrerence in my experience.
DC 12V-24V TDA7492MV Mono 50W Class D Audio Power Amplifier Board Amp Module 699971085729 | eBay
and looked up the datasheet for the TDA7492MV chip from ST. The 50 watts claimed for the eBay board is ********, of course, but the datasheet indicates a fairly clean (1% THD) 12 watts into 8Ω with a 15 volt supply. That'll do me. Then my buddy found these:
4X 18650 3.7V 9800mAh TR Li-ion Rechargeable GTF Battery + Smart US Charger OY | eBay
8 of them in series-parallel would give me ~ 15 volts at 2 A-H and only cost about $16. Hmmmmm......
I'll end up spending more on the speaker than all the rest, but that is only right, methinks, because it makes the most diffrerence in my experience.
Last edited:
The ad copy is difficult to parse, but my guess is that the "3.7V 9800 mAh" is for for cells in parallel. That would make each cell 2450 mAh, or 2.45 Ah, which sounds about right for cells of that physical size.
The price is great, but:
1) These are not LiFePO4, and have a correspondingly increased risk of fire. Only you can decide whether you are willing to risk having such Lipo / Lion packs in your home.
2) There is no balancer circuit provided.
Charging cells in parallel eliminates the need for a balancer - but you have to put them in series to get the voltage you need. So, if you charge in parallel, you need some kind of a 4-cell battery holder for 18650-sized cells.
(Most lithium packs have cells soldered or spot-welded permanently together, rather than individually charged and dropped into a battery holder the way we use AA cells.)
-Gnobuddy
Speaking of Li powered drills and the like that get charged in the home all the time. Balancing and protection. Just need a charger. 5S 18V 20A PCM PCB Li-ion Lithium Battery Protect Circuit Board For Makita Drill 8784253970410 | eBay
There are currently 3 common lithium battery chemistries in use today. They are NOT compatible with each other. Using the wrong controller board or charger can lead to disaster.
Lithium Ion (LiIon or Lion) was the original lithium chemistry used in phones and laptops. It is reasonably safe as long as proper techniques are used. Nominal cell voltage is 3.6 or 3.7 volts, although a few higher voltage cells were seen. They all have a minimum discharge voltage. The cells will be irreversibly damaged if discharged below this voltage, and any apparatus using these cells should self disconnect before the battery reaches this voltage. There are also maximum cell voltages and temperatures that must be respected. Cell to cell balance is usually used, but not always required in light discharge applications. Most lithium ion cells are encased in metal. It has a reasonably energy density, high energy to volume ratio, medium energy to weight ratio.
Lithium Polymer (LiPo or Lipo) is the cells commonly used in cell phones and laptops today. We evaluated this tech for our phones in the late 90's and deemed it unsafe. The transition happened after I left the phone design world (around 2005). This is the chemistry seen in virtually all the flaming phone and laptop incidents. The cells have a nominal voltage of 4.0 to 4.2 volts and share all of the temperamental characteristics of Lion, with flaming consequences if not followed. Cell to cell balance is mandatory or one cell could be discharged too far, then flame up due to the action of the charge / discharge current from the other cells. All modern cell phones are single celled, so there is no balance required. The cells are usually encased in thin plastic. Fire will usually happen if the casing is penetrated and air or moisture enters.
The battery packs used by the model helicopter / drone hobby are multi celled LiPo batteries. A specific balancing charger is required. I have one, and it is far too easy to select the wrong chemistry with possible ugly consequences. 22 volt 5500 mAH (over 100 WH) packs were $50 when I got them 4 years ago. They still test out at over 4000 mAH today, but I rarely use them any more. They are stored in a thick metal ammo box which is kept on concrete several feet away from any wood. These have the highest energy density of all currently used battery chemistry, but the worst safety record.
LiFePo4 (lithium iron phosphate) These are the safest of the 3 common lithium chemistries, but the heaviest. Nominal cell voltage is 3.2 volts. A cell slightly longer than a "D" cell can have 5200 mAH. Again there are voltage ranges that must be observed on charge and discharge cycles. Cell balancing is best for extracting all of the capacity, but 80 to 90% can be realized by float charging.
This is what I use in my portable computer. There 7 of these cells wired in series for 22.4 volts float charged from a 26 volt supply with a diode in series. Maximum charge voltage is 3.6 to 3.8 volts per cell. I stick to 3.6 volts on a float charge. Measured cell current if left plugged in drops to 1 or 2 mA, but cells don't get a 100% charge this way. I use a mechanical relay to fully disconnect cells on power down since the PC can go for months without use.
I may look into a balance charger, but it's not high on my list. The core I7-7700T computer with a 20 inch LCD screen gets about 4 hours of average use per charge, and about 1.5 hours maxed out running 8 threads (2 per core) of Prime 95. I plan on rebuilding it into a 4 K "laptop" with a 15.6 inch screen that consumes less power.
For a guitar amp that is already going to be somewhat heavy, I would use LiFePo4 cells, with or without a balance board. The balance boards shown here are for Lion (3.7 volt) cells. They are not compatible with LiFePo4 (3.2 volt) cells without modifications.
Lithium Ion (LiIon or Lion) was the original lithium chemistry used in phones and laptops. It is reasonably safe as long as proper techniques are used. Nominal cell voltage is 3.6 or 3.7 volts, although a few higher voltage cells were seen. They all have a minimum discharge voltage. The cells will be irreversibly damaged if discharged below this voltage, and any apparatus using these cells should self disconnect before the battery reaches this voltage. There are also maximum cell voltages and temperatures that must be respected. Cell to cell balance is usually used, but not always required in light discharge applications. Most lithium ion cells are encased in metal. It has a reasonably energy density, high energy to volume ratio, medium energy to weight ratio.
Lithium Polymer (LiPo or Lipo) is the cells commonly used in cell phones and laptops today. We evaluated this tech for our phones in the late 90's and deemed it unsafe. The transition happened after I left the phone design world (around 2005). This is the chemistry seen in virtually all the flaming phone and laptop incidents. The cells have a nominal voltage of 4.0 to 4.2 volts and share all of the temperamental characteristics of Lion, with flaming consequences if not followed. Cell to cell balance is mandatory or one cell could be discharged too far, then flame up due to the action of the charge / discharge current from the other cells. All modern cell phones are single celled, so there is no balance required. The cells are usually encased in thin plastic. Fire will usually happen if the casing is penetrated and air or moisture enters.
The battery packs used by the model helicopter / drone hobby are multi celled LiPo batteries. A specific balancing charger is required. I have one, and it is far too easy to select the wrong chemistry with possible ugly consequences. 22 volt 5500 mAH (over 100 WH) packs were $50 when I got them 4 years ago. They still test out at over 4000 mAH today, but I rarely use them any more. They are stored in a thick metal ammo box which is kept on concrete several feet away from any wood. These have the highest energy density of all currently used battery chemistry, but the worst safety record.
LiFePo4 (lithium iron phosphate) These are the safest of the 3 common lithium chemistries, but the heaviest. Nominal cell voltage is 3.2 volts. A cell slightly longer than a "D" cell can have 5200 mAH. Again there are voltage ranges that must be observed on charge and discharge cycles. Cell balancing is best for extracting all of the capacity, but 80 to 90% can be realized by float charging.
This is what I use in my portable computer. There 7 of these cells wired in series for 22.4 volts float charged from a 26 volt supply with a diode in series. Maximum charge voltage is 3.6 to 3.8 volts per cell. I stick to 3.6 volts on a float charge. Measured cell current if left plugged in drops to 1 or 2 mA, but cells don't get a 100% charge this way. I use a mechanical relay to fully disconnect cells on power down since the PC can go for months without use.
I may look into a balance charger, but it's not high on my list. The core I7-7700T computer with a 20 inch LCD screen gets about 4 hours of average use per charge, and about 1.5 hours maxed out running 8 threads (2 per core) of Prime 95. I plan on rebuilding it into a 4 K "laptop" with a 15.6 inch screen that consumes less power.
For a guitar amp that is already going to be somewhat heavy, I would use LiFePo4 cells, with or without a balance board. The balance boards shown here are for Lion (3.7 volt) cells. They are not compatible with LiFePo4 (3.2 volt) cells without modifications.
Thanks for chiming in with your expertise, George!
From what I've seen, lipo pack manufacturers typically specify a maximum storage temperature of 150 degrees Fahrenheit / 65.5 degrees Celsius. Spontaneous ignition may occur if the temperature goes above 170 deg. F / 77 deg C. ( LiPo Storage )
According to the US Centers for Disease Control and Prevention, "When temperatures outside range from 80 degrees to 100 degrees, the temperature inside a car parked in direct sunlight can quickly climb to between 130 to 172 degrees" ( Hospital CEO Leaves Child to Die in Hot Car - ABC News )
Charging was done outdoors in the cavity of a cement breeze-block (concrete brick), with the brick placed on a concrete paving-stone, and nothing flammable for at least ten feet / three metres in every direction, including straight up.
When a lipo pack goes up in flames, the result is similar to a Roman Candle (firework). There is no explosion - but there is an eruption of flames and a spray of high-temperature particles, accompanied by choking clouds of very unpleasant-smelling thick black smoke.
The spray of high-temperature sparks and flaming fragments can set fire to any combustible materials in the vicinity (such as soft furnishings and car interiors.) On concrete or bare earth, there is little danger.
I used to use these with my LiFePO4 packs: 106-123 Astro "Blinky" Battery Balancer for A123 Cells
-Gnobuddy
One more dangerous failure mode - storage at too high a temperature also turns out to have flaming consequences with these lipo cells. The internal chemistry goes unstable at temperatures that are easily reached on a vehicle dashboard in warm sunny climates, or inside a model aircraft with a clear plastic canopy in Arizona, where I first heard about the issue.
From what I've seen, lipo pack manufacturers typically specify a maximum storage temperature of 150 degrees Fahrenheit / 65.5 degrees Celsius. Spontaneous ignition may occur if the temperature goes above 170 deg. F / 77 deg C. ( LiPo Storage )
According to the US Centers for Disease Control and Prevention, "When temperatures outside range from 80 degrees to 100 degrees, the temperature inside a car parked in direct sunlight can quickly climb to between 130 to 172 degrees" ( Hospital CEO Leaves Child to Die in Hot Car - ABC News )
I took similar precautions with these, as well as charging and storing with Ziplock bags full of sand on top and around the pack. (In case of fire, the bag burns through, and the sand contains the eruption of flaming debris as well as smothering the fire.)
Charging was done outdoors in the cavity of a cement breeze-block (concrete brick), with the brick placed on a concrete paving-stone, and nothing flammable for at least ten feet / three metres in every direction, including straight up.
When a lipo pack goes up in flames, the result is similar to a Roman Candle (firework). There is no explosion - but there is an eruption of flames and a spray of high-temperature particles, accompanied by choking clouds of very unpleasant-smelling thick black smoke.
The spray of high-temperature sparks and flaming fragments can set fire to any combustible materials in the vicinity (such as soft furnishings and car interiors.) On concrete or bare earth, there is little danger.
True, except for the one 4S LiFePO4 pack that Dotneck linked to, which came with its own built-in balancer and protection circuit. One hopes those were tailored to the LiFePO4 chemistry they were being used with.
I used to use these with my LiFePO4 packs: 106-123 Astro "Blinky" Battery Balancer for A123 Cells
-Gnobuddy
????????? "The Astro 123 Blinky is specially designed for A123 Lithium Ion Batteries."
Technically, I think it's true that all lithium batteries actually depend on lithium ions in the electrolyte to transport the current...
The ad copy may be confusing (perhaps cut and pasted from the companys advertising for their Blinky balancers for regular Lipo packs.) But A123 Systems only made LiFePO4 cells.
The company (A123 Systems) was spun off from research into nanoparticle anodes (which increased electrode surface area) at MIT, and were the first LiFePO4 cells I know of that could be charged in 15 minutes, and discharged at more than 10 C (a 2.3 Ah cell could be discharged at 35 amps with no drama.)
Lithium battery terminology seems to leave a lot to be desired. From my reading, it appears that todays Lipo and Li-ion batteries actually use the same chemistry, and the only real difference is that the li-ion ones roll up the electrodes tightly and stuff them into a metal can for protection, while Lipo packs use the equivalent of a tiny Ziplock bag to contain the electrodes and electrolyte:
"What is the difference between a normal Li ion and Li ion polymer? As far as the user is concerned, lithium polymer is essentially the same as lithium-ion. Both systems use identical cathode and anode material and contain a similar amount of electrolyte." (That excerpt was taken from here: Li-polymer Battery: Substance or Hype? – Battery University )
Equally confusingly, since both are actually the same chemistry, both Lipo and Li-ion actually charge up to 4.2V and droop to maybe 3.5 V fully discharged, but Lipo packs are often advertised at 4.2 V/cell, while li-ion cells are typically rated at 3.7 V/cell.
DeWalt has a line of lithium-powered cordless tools advertised with the trade name "20V Max" ( 20V MAX* Cordless Tools & Batteries | DEWALT ). The packs have 5 li-ion cells in series - so apparently DeWalt is one of the few li-ion users who recognize that li-ion cells actually produce (a bit more than) 4 volts per cell, fully charged, and also that boasting about 20V while the competition says "18.5 V" might get you a few extra sales from the uninformed.
One of the reasons I'm cautious about the supposedly safer li-ion cells is the fact that they are in fact the same thing as lipo cells, except for the better protection from impact damage or rupture provided by the rigid metal case. Since there are many Lipo failure modes that do not involve physical impact or rupture of the outer envelope, I wonder how much safer Li-ion cells actually are.
LiFePO4, on the other hand, is safer because of the nature of the internal chemistry, and not because of the outer housing.
-Gnobuddy
"What is the best eBay special Class-D amp that will do an honest 12 watts into 8Ω from a 12 volt supply???"
Well, to answer my own question, I think that's not quite achievable. Looking at the TDA7492MV datasheet, the best that's possible from a 12 volt supply is ~ 6 watts into 8Ω@ 1% THD; 8 watts into 8Ω @ 10% THD. That's not quite up to a Princeton, but perhaps close enough. Does anyone know of another Class-D chip that will do better?
Well, to answer my own question, I think that's not quite achievable. Looking at the TDA7492MV datasheet, the best that's possible from a 12 volt supply is ~ 6 watts into 8Ω@ 1% THD; 8 watts into 8Ω @ 10% THD. That's not quite up to a Princeton, but perhaps close enough. Does anyone know of another Class-D chip that will do better?
Page 10, http://www.ti.com/lit/ds/symlink/tpa3116d2.pdf
You are going to get 10W @ 10% distortion with an 8 ohm load. Won't matter too much what chip you use, Ohms law isn't going to be bent much. Peak at 12V means 12*0.7 = 8.4V RMS, 8.4/8 ohms = 1A, 1A*8.4 = 8.4W, need more voltage.
That is why I suggested the voltage multiplier. 3A DC-DC Boost Step-up Converter Voltage Regulator 5V-35V to 9V 12V 24V 36V 48V | eBay
15W at 16V is doable, need 50% more current though. Or use a 4 ohm speaker.
You are going to get 10W @ 10% distortion with an 8 ohm load. Won't matter too much what chip you use, Ohms law isn't going to be bent much. Peak at 12V means 12*0.7 = 8.4V RMS, 8.4/8 ohms = 1A, 1A*8.4 = 8.4W, need more voltage.
That is why I suggested the voltage multiplier. 3A DC-DC Boost Step-up Converter Voltage Regulator 5V-35V to 9V 12V 24V 36V 48V | eBay
15W at 16V is doable, need 50% more current though. Or use a 4 ohm speaker.
Pretty much what I was going to say. If you go LiFePO4, six cells in series (around 18 - 19 volts) is what you need.
There isn't much penalty for having a bit more voltage, so if your class D board will handle it, two of those 4-cell LiFePO4 packs (the ones you linked to earlier) in series (for 25 - 26 volts total) may be the easy way to go.
Class D amps, in my experience, clip very, very unpleasantly. It wouldn't hurt to have a little extra voltage headroom by using an 8 cell LiFePO4 pack.
-Gnobuddy
Speaking of (everybody says were we?) I put a Blackface tonestack between my Mu-amps feeding the Class D amp. Enough gain to overdrive the second stage a bit, way more than enough drive to get the Class D to full output. The next step would be to see how much signal voltage it needs and to clip or limit the voltage to the amp.
7 cells require 25.2 volts to reach a 90% charge (3.6 volts per cell) without a balance circuit. Add .3 volts drop across a Shottkey diode gives 25.5 volts. This is about the max I could get from a Meanwell 24 volt 150 watt SMPS (RS-150-24). 8 cells would need a larger charging voltage, but prewired packs are available. I stuffed the cells into a piece of PVC pipe with a spring in one end cap, and a bolt in the other to make a DIY battery holder. The end cap is glued at one end, slotted and hose clamped at the other for cell removal. It's all been working for about 9 months now.
I need 100 watts to run the PC at full tilt if the batteries are not helping. The 150 watt supply was actually cheaper than a smaller supply when I got it from Jameco. 150 watts supplies enough grunt to charge batteries and run the PC, and the class D amp inside it.
I am using a small class D amp from the PC's 12 volt regulator. It provides more power than the little speakers can eat. It's only good for 4 WPC or so.
Any class D amp that uses a TPA3116 chip should be able to eat 24 to 26 volts and provide "50 WPC." TI's data sheet provides the 50 watt number, it's at a 4 ohm load and 10% THD. These boards sound good for HiFi and guitar provided they are kept away from clipping. Some cheap boards might use 25 volt caps though.
The power VS supply voltage for 1% THD curve with an 8 ohm load in the data sheet reveals about 23 WPC at 20 volts, 27 WPC at 22 volts, 33 WPC at 24 volts, and 37 WPC at 26 volts. I haven't measured mine but they are really loud with a 4 ohm load on each channel while fed from a 24 volt SMPS. The 150 W MeanWell will support it at full crank without issue. My board was a generic Chinese unit from Banggood. Parts Express has TPA3116 boards from $9 to $25. Amazon has them too. Just stick TPA3116 into either companies search engine.
I need 100 watts to run the PC at full tilt if the batteries are not helping. The 150 watt supply was actually cheaper than a smaller supply when I got it from Jameco. 150 watts supplies enough grunt to charge batteries and run the PC, and the class D amp inside it.
I am using a small class D amp from the PC's 12 volt regulator. It provides more power than the little speakers can eat. It's only good for 4 WPC or so.
Any class D amp that uses a TPA3116 chip should be able to eat 24 to 26 volts and provide "50 WPC." TI's data sheet provides the 50 watt number, it's at a 4 ohm load and 10% THD. These boards sound good for HiFi and guitar provided they are kept away from clipping. Some cheap boards might use 25 volt caps though.
The power VS supply voltage for 1% THD curve with an 8 ohm load in the data sheet reveals about 23 WPC at 20 volts, 27 WPC at 22 volts, 33 WPC at 24 volts, and 37 WPC at 26 volts. I haven't measured mine but they are really loud with a 4 ohm load on each channel while fed from a 24 volt SMPS. The 150 W MeanWell will support it at full crank without issue. My board was a generic Chinese unit from Banggood. Parts Express has TPA3116 boards from $9 to $25. Amazon has them too. Just stick TPA3116 into either companies search engine.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.