Certainly the primary bulk cap cannot help deliver beyond the current limit, which, in any case, should not be exceeded during all kind of operation.
Remember, you have plenty of loop gain (feedback) at lower frequencies to reduce the effects of PSU ripple, so getting rid of higher frequencies is really the important thing as you lose feedback.
Using say C1 = Cd = 2200 uF, something like this would work, but I'd recommend a damping resistor (Rd) of ~ 0.05 ohms (parallel 0.1 ohm resistors and you want to have some oomph) to damp the Q:
DENO-23-0001 Schurter | Mouser
Just to give an idea at least. Probably excessive! Also, not sure how much it's going to help above say 100 kHz where self-resonance and ESR of the respective caps is going to kick in hard.
Last edited:
Actually I'm sitting here scratching my head as the inductor I linked doesn't add up. It would make sense as a 2 uH inductor, but definitely not a 2 mH. (Schurter using m as micro?)
In any case, something like a 330 uH (0.33 mH) choke with a 2200 uF primary capacitor and a damping cap of an additional 2200 uF +series ~0.7 ohm resistor *should* play nice. And not be outrageous.
In any case, something like a 330 uH (0.33 mH) choke with a 2200 uF primary capacitor and a damping cap of an additional 2200 uF +series ~0.7 ohm resistor *should* play nice. And not be outrageous.
Using:
Designing Second Stage Output Filters for Switching Power Supplies | Analog Devices
Technique 2, Figure 4 is what I'm suggesting.
Designing Second Stage Output Filters for Switching Power Supplies | Analog Devices
Technique 2, Figure 4 is what I'm suggesting.
A couple of the 258rmb Taobao boards arrived yesterday evening - and I started looking at the power supplies for the signal stages (input opamps and 12V to the TPA3255). A quick listen first and that was promising, the usual classD clarity of bass lines - far from transparent in the HF though.
Checking the DS for the LM2575 (note not -HV so supplies above 40V won't be tolerated) and looks like the board's designer didn't take to heart the guidelines expounded there. The output inductor for this (buck) reg is marked on the silkscreen as '220uH' but what's fitted is a 22uH (220 marking). My reading for the nomographs in the DS (fig 30 pertains as its a -15) shows a recommended inductor of 2.2mH, which I just happen to have in my box.
The output capacitor fitted is a 10uF/35V SMT. I removed this and checked its ESR - around 5ohms. To replace it I went with a Sancon 220uF with about 0.1ohm ESR. Note the DS cautions against caps here with ESR below 0.05ohm - stability may suffer.
I've not 'scoped this yet but I figure having the correct inductor value (100X bigger than fitted) and an output cap 22X larger has to help with output ripple.
Next stage - an LC on the output.
1st project - looking at TPA3255
Attempting Large, just portable, bluetooth battery powered speakers.
Would it be best to go 4P3S with 18650's and get 48V with something like this ?
Or just go 11S 46.2Vmax with a 10A fuse ?
Or some other way ?
Advice very much appreciated.
PS. I have a 6 cell balanced charger so maybe it would be sensible to go 2P6S and just run it at 25.2Vmax
Would I be able to get about 2x30W at low distortion like this ?
Attempting Large, just portable, bluetooth battery powered speakers.
Would it be best to go 4P3S with 18650's and get 48V with something like this ?
Or just go 11S 46.2Vmax with a 10A fuse ?
Or some other way ?
Advice very much appreciated.
PS. I have a 6 cell balanced charger so maybe it would be sensible to go 2P6S and just run it at 25.2Vmax
Would I be able to get about 2x30W at low distortion like this ?
Last edited:
Ooops, batteries I have are max 4.8A so 11S not a great idea ...or use with a 4A fuse?
Hi, I'm new to the forum.
My TDA3255EVM will be shipped in a week or so, can't wait!
The only PSU I can get for free is a Siemens Gleichrichter GR 1500W Rectifier (S30050-K6383-x-1), label says 48(54)VDC and 16(14)/31(28)A on the output stage.
What do you guys think? Overkill? Hell yes!! But would it work nicely with the TPA3255EVM?
I will only drive bookshelf speakers for now. Ironic since I dont like to listen to music too loudly (like quality over quantity), but I have the biggest PSU on the forum?
My TDA3255EVM will be shipped in a week or so, can't wait!
The only PSU I can get for free is a Siemens Gleichrichter GR 1500W Rectifier (S30050-K6383-x-1), label says 48(54)VDC and 16(14)/31(28)A on the output stage.
What do you guys think? Overkill? Hell yes!! But would it work nicely with the TPA3255EVM?
I will only drive bookshelf speakers for now. Ironic since I dont like to listen to music too loudly (like quality over quantity), but I have the biggest PSU on the forum?
Yessir, overkill. To drive bookshelf speakers not too loudly, any TPA3116 board powered by a notebook adapter will do the trick.
Found a pic of the power supply and it looks like it has a fan. That's an industrial-strength, high-reliability telecom power supply...and while I have no experience with it, it is probably not a good choice for 2 reasons:
1. Fan probably too loud to use in your listening room
2. Will use a lot of watts from the wall just to have the power supply on, WAY more than you need, and that means heat dissipation. My most efficient psu/tpa3255evm combination (400W psu) drives the tpa3255evm using only ~7W from the wall at idle and only a tiny bit more while in use. Guessing that GR90 psu will probably use 30W+ at idle, but can't find specs.
Yeah, fan is too loud, but I will play with it, maybe remove fan and the internal heatsink (looks like it could be removed very easily), exposing the regulators to be easily mounted to the case heatsink for silent operation.
As for idle usage, only way would be to measure it, because Google cannot find the secret hidden specs. Might not be that much, seems to be very "intelligent"?
As for idle usage, only way would be to measure it, because Google cannot find the secret hidden specs. Might not be that much, seems to be very "intelligent"?
