So that is just 10W average at 'war level' (sound volume); that is great.
How high are the peaks the batteries need to deliver at that volume?
Running it at 36V probably helps quite a bit with less idle power.
Btw, are your SLA's in series or parallel?
30-80% is very conservative to get many more cycles out of them; I regularly (undercharge) my 10S LIPO ebike pack to 40.5V; (2 diodes @charger plug:4x more cycles), that should be ~90% as far as I remember correctly.
About every 20-30 charges, I charge to 42V to balance the cells with a full charge and let the bms work, but I 'm still not sure if this is necessary. Do you?
These 10s2p li ion packs can be run down in about half an hour and the 2x250w electric motors have peak draws a few times higher; so running an amp with 10W average should be peanuts and run for many hours.
Have you tested the fake cells/pack? Because if they perform much worse then real/original, the toys who use them would not be usable. But they might still be fine to run an amp for hours.
How high are the peaks the batteries need to deliver at that volume?
Running it at 36V probably helps quite a bit with less idle power.
Btw, are your SLA's in series or parallel?
30-80% is very conservative to get many more cycles out of them; I regularly (undercharge) my 10S LIPO ebike pack to 40.5V; (2 diodes @charger plug:4x more cycles), that should be ~90% as far as I remember correctly.
About every 20-30 charges, I charge to 42V to balance the cells with a full charge and let the bms work, but I 'm still not sure if this is necessary. Do you?
These 10s2p li ion packs can be run down in about half an hour and the 2x250w electric motors have peak draws a few times higher; so running an amp with 10W average should be peanuts and run for many hours.
Have you tested the fake cells/pack? Because if they perform much worse then real/original, the toys who use them would not be usable. But they might still be fine to run an amp for hours.
An update for my evm board and voltage booster disaster:
The shorted voltage booster, rated 400W 8A cont and 15A peak, max 60V, has a shorted IRFB3607 mosfet which is rated for 144W max......
Expensive POS......
So what probably happened is that the mosfet died, stayed open, voltage rose till I got the spark to the heatsink which melted the 6 power pins.
Still the mosfet should not have died as max I have seen going into the board was a bit less then 1A @48V
Any input still welcome..
While soldering the short ends sticking out of the chip back to the board, I had to give the left side a few touches more then I liked.
Solder job looked pretty good, but when powering it with 20v the left part of the chip burned and now I had the familiar smell..... I will order a new chip.
The shorted voltage booster, rated 400W 8A cont and 15A peak, max 60V, has a shorted IRFB3607 mosfet which is rated for 144W max......
Expensive POS......So what probably happened is that the mosfet died, stayed open, voltage rose till I got the spark to the heatsink which melted the 6 power pins.
Still the mosfet should not have died as max I have seen going into the board was a bit less then 1A @48V
Any input still welcome..
While soldering the short ends sticking out of the chip back to the board, I had to give the left side a few touches more then I liked.
Solder job looked pretty good, but when powering it with 20v the left part of the chip burned and now I had the familiar smell..... I will order a new chip.
Last edited:
There’s nothing wrong about the FET ratings, power loss in switched applications is I^2*Rdson + switching losses, so even at 50A input + ripple current Pd is lower than 25W. (Which of course is way to much without über cooling.) To stay below 3W Pd, max current needs to be <=18A. Taking Rdson increase into account, assuming x2, so it is 18mR then, current must be smaller than 13A.
Their might be a thermal runaway situation due to collapsing magnetic flux in the inductor which then fried the FET with over current and overheating. Keep in mind that these current limits are meant for input side. Inductor ripple current increases dramatically when flux is reduced.
Their might be a thermal runaway situation due to collapsing magnetic flux in the inductor which then fried the FET with over current and overheating. Keep in mind that these current limits are meant for input side. Inductor ripple current increases dramatically when flux is reduced.
Thank you.
IRFB3607 Data sheet https://www.infineon.com/dgdl/irfs3607pbf.pdf?fileId=5546d462533600a401535636a42b2176
Sorry my fault; the data sheet says 140W max power dissipation, (not max power) so indeed that should not be the problem as the heat sinks made good contact and were not even warm to the touch.
Btw it is this booster 8.5-50V to 10-60V 400W DC DC Boost Converter Charger Module for Car/Notebook | eBay an I have a second one which I have not used yet.
Caps are 63V 220uF and seem to be fine. Diode is the one without a heat pad in between it and it's sink.
I can't stand that I don't understand what happened and I guess I have to take my electro skills to a higher level before I do.
Is there a chance that the arcing at the chip killed the mosfet on the voltage booster?
IRFB3607 Data sheet https://www.infineon.com/dgdl/irfs3607pbf.pdf?fileId=5546d462533600a401535636a42b2176
Sorry my fault; the data sheet says 140W max power dissipation, (not max power) so indeed that should not be the problem as the heat sinks made good contact and were not even warm to the touch.
Btw it is this booster 8.5-50V to 10-60V 400W DC DC Boost Converter Charger Module for Car/Notebook | eBay an I have a second one which I have not used yet.
Caps are 63V 220uF and seem to be fine. Diode is the one without a heat pad in between it and it's sink.
I can't stand that I don't understand what happened and I guess I have to take my electro skills to a higher level before I do.
Is there a chance that the arcing at the chip killed the mosfet on the voltage booster?
The tops are around 5R connected in SE. That's (50V*0.9*0.707/2)^2/5R=50Wrms -> 100W peak
The sub is 4R in BTL, which renders to: (50V*0.9*0.707)^2/4R=250Wrms -> 500Wpeak
Worst case would be 500W+2*100W = 700Wpeak
Considering a 0.85 efficiency for the chinese boost converter, there's 700W/0.85= 825W on the input. With 36V system voltage, current peaks are at 23A.
Series. The advantage of SLA is their low impedance compared to LiIon-packs. a 36V SLA string mostly under 60-70mR, where a LiIon pack, even with 4 cells in parallel easily hits 100-150mR.
Balancing from time to time isn't anything bad, depends on cell selection. Best practice is to store them away at about 30-40%, so discharging after use helps to slow down aging alot.
I have tested them, the whole pack only had 70% of "new" capacity when bought. Beside this, cells and pcb showed rust and signs of use.
140W is textbook and a calculated value as it is impossible to keep the silicon at 25°C while pushing max current trough..
Btw it is this booster 8.5-50V to 10-60V 400W DC DC Boost Converter Charger Module for Car/Notebook | eBay an I have a second one which I have not used yet.
Caps are 63V 220uF and seem to be fine. Diode is the one without a heat pad in between it and it's sink.
I can't stand that I don't understand what happened and I guess I have to take my electro skills to a higher level before I do.
Is there a chance that the arcing at the chip killed the mosfet on the voltage booster?
I suspect the converter failed due to thermal runaway problem. When these toroids heat up, they'll loose their magnetic quite instantly when pushed hard. Not only that, they also shift their "breakeven point" to lower currents to.
I actually use one of these 1200W converters (20A input current):
1200W 20A DC Converter Boost Car Step-up Power Supply Module 8-60V to 12-83V 2 | eBay
There's also a 1500W model available with fan + doubled toroid (30A input current):
1500W 30A Voltage Step Up Converter Boost CC CV Power Supply Module MB 8307768981315 | eBay
A cheaper solution is those 600W converters, which are available for a few years now.. (10A input current limit)
600W DC-DC 10-60V to 12-80V Boost Converter Step-up Module Power Supply 701413409670 | eBay
I'd recommend the 1500W version, the fan isn't needed normally.
Last edited:
Hi, mates. Uninstalled hetasink to show the chip topology. Got questions: 1) how do comment the topology?
2) main capasitor is only 1000/50V without preventing cap.is it ok? In description seller noticed 2200/63V elna.
3) saw some thermopaste in-between chips legs, is it ok? Have evacuated some amount by tampon. Any suggestions are welcome!
2) main capasitor is only 1000/50V without preventing cap.is it ok? In description seller noticed 2200/63V elna.
3) saw some thermopaste in-between chips legs, is it ok? Have evacuated some amount by tampon. Any suggestions are welcome!
Attachments
Hi Antony
1; BTL mode
2;the caps in Class D must be with low ESR/ESL and normaly the chinese board provide a "normal" cap...lets say old version and long storage time because the buy it in 100.000 peaces..
if you use 48V you should use the next Voltage value e.g. --> 63V to have some headroom...etc..
datasheet TPA3255EVM -
the Caps 4700µF/ 80V should be connect to the Rails of the Power chip .
PVDD (BTW- ind the TPA3255 chip datasheet there is wrong written=470µF)
look at the attachment
3; they use too much....., if you have an PCB cleaner you can clean it and use a newer and less past to contact here....
chris
Attachments
Originally Posted by AnatoliyHill View Post
Wow, Dear Think mentioned original topology contain 2#4700mkF which almost double more I have in my box. Is this a main reason for lack of bass I am having? Furthermore I have only one capacitor enstead of two, think it worth to take of sink enable mates see the topology for further decision.
doc ---> As written already - no.
hi anatoliy
Doc is principal right about bass and µF on the rails.....but in your second pic(complete board) i try to check the pcb traces ....its not sharp....and it looks like that the blue cap 220µF/50V is connected to PVDD_CD pin 31,30, 29. so its not according to the datahseet!!!
and it slooks like that the trace also go from PVDD_AB pin 38,37,36 ! so its 1 cap for both rails?!!!!
maybe somebody with better eyes can confirm this
...but with your "loudness" or volume concern it has nothing to do
chris
Wow, Dear Think mentioned original topology contain 2#4700mkF which almost double more I have in my box. Is this a main reason for lack of bass I am having? Furthermore I have only one capacitor enstead of two, think it worth to take of sink enable mates see the topology for further decision.
doc ---> As written already - no.
hi anatoliy
Doc is principal right about bass and µF on the rails.....but in your second pic(complete board) i try to check the pcb traces ....its not sharp....and it looks like that the blue cap 220µF/50V is connected to PVDD_CD pin 31,30, 29. so its not according to the datahseet!!!
and it slooks like that the trace also go from PVDD_AB pin 38,37,36 ! so its 1 cap for both rails?!!!!
maybe somebody with better eyes can confirm this
...but with your "loudness" or volume concern it has nothing to do
chris
@ AnatoliyHill
Your Amplifier is the one i suggested?
Cheap Aliexpress TPA3255 amplifier
It looks nice. I dont receive mine at the moment.
PCB looks not too bad ^^
And why is TPA3250 not much cheaper. It has no bluetooth and less power.
Cheap TPA3250 amplifier
I hope also that in the near future they will release a TPA3251 amplifier. It should be the best compromise between power and price :-D
Your Amplifier is the one i suggested?
Cheap Aliexpress TPA3255 amplifier
It looks nice. I dont receive mine at the moment.
PCB looks not too bad ^^
And why is TPA3250 not much cheaper. It has no bluetooth and less power.
Cheap TPA3250 amplifier
I hope also that in the near future they will release a TPA3251 amplifier. It should be the best compromise between power and price :-D
Last edited:
Chris, you are right hereis only one capasitor 50v/220mkF think work got both channel s. Yeap, this is one of cheap tpa3255 . Think I should go for some tpa3255 which more close to original topology like yours in order to get more solid sound. The only reason I bought this cheap one device because it is finished and with BT. This tpa3255 is producing sound power similar to tpa3116 with 18v/4A laptop adapter - my personal opinion. Can you please provide link for reliable seller from AliExpress?
hi anatoliy
Why not modding this amp? its not bad and i understand your statements to buy it.
for me its important to clarify if its in SE or BTL mode...i am not really sure. the pic are not clear. espacially at the output filter coils and caps...this is important because SE mode is not the best solution for good SQ. (i get a comment by Voltwide about this)
lets mod this amp.
please try to get a sharp pic from the ouputfilter section...hopefully we get help from others to read the traces on the pcb
hers is a link...but sears which shop is better to deliver in your country..
1 STUCK Neue Ankunft Klasse D TPA3255 MINI HIFI AUDIO Digital Verstarker brett 300 Watt + 300 Watt DC50V Modul in 1 STUCK Neue Ankunft Klasse D TPA3255 MINI HIFI AUDIO Digital Verstarker-brett 300 Watt + 300 Watt DC50V Modul aus Integrierte Schaltungen auf AliExpress.com | Alibaba Group
chris
Thanks, Chris. German store is not available for order from Ukraine. Will try from this source, board is looking similar
1 401,56 грн. 7%СКИДКА | Высокое качество класса D TPA3255 HIFI аудио цифровой усилитель совета разрядный Мощность схема усилителя модуля 300 Вт + 300 Вт DC 48 В
High Quality Class D TPA3255 HIFI AUDIO Digital Amplifier Board Digit Power AMP Circuit Module 300W + 300W DC 48V -in Amplifier from Consumer Electronics on Aliexpress.com | Alibaba Group.
Will orovide more pictures for further moding. Meanwhile I am pretty sure my board is extreme cheap interpretation of tpa3255 original topology , however is on acceptable level of versatility and sound. The only problem to handle canron sc-l with DIN 200W it not enough.
1 401,56 грн. 7%СКИДКА | Высокое качество класса D TPA3255 HIFI аудио цифровой усилитель совета разрядный Мощность схема усилителя модуля 300 Вт + 300 Вт DC 48 В
High Quality Class D TPA3255 HIFI AUDIO Digital Amplifier Board Digit Power AMP Circuit Module 300W + 300W DC 48V -in Amplifier from Consumer Electronics on Aliexpress.com | Alibaba Group.
Will orovide more pictures for further moding. Meanwhile I am pretty sure my board is extreme cheap interpretation of tpa3255 original topology , however is on acceptable level of versatility and sound. The only problem to handle canron sc-l with DIN 200W it not enough.
I found it a bit cheaper.
I am waititng for my 500w LLC supply to drive my IRS2092 amp. But my temptation is big to buy a TPA3255 for testing...
2.0 Digital amplifier board DC50V Strong high power TPA3255 300W + 300W Class D audio Digital Amplifier Board
I am waititng for my 500w LLC supply to drive my IRS2092 amp. But my temptation is big to buy a TPA3255 for testing...
2.0 Digital amplifier board DC50V Strong high power TPA3255 300W + 300W Class D audio Digital Amplifier Board
canron sc-l with DIN 200W it not enough
???
all the chassis in the speaker an max 200W but this is for sure not in the linear movement of the speakers...so a lot of distortions
i did some mearuement on real volume level in livingroom its about 1-15W (max 30-50W) per channel . so good sound is not a task of just of power.
let compare some worlds:
the amp/speaker is the "same relation" like car/engine and tires
a small car can use a tire with max speed is allowed 230km/h
a big car can handle also a tire with Vmax 180...but you have to be carefully not go over the limit
chris
Yep, not enough with this device. I mean it is sounds with no distortion even at max level, bass speakers is having not too much excursions and well controled. No wonder at 8Homs tpa3255 produce only 125W. I know how canton sc-l can sounds with 100w amplifier AB class on 8 hom moving whole air in the room.