Interesting
Did you take in account the peculiar LPF needed by AK4396 (fc=125kHz, Q=0.692)?
So, probably, FCs are not causing overheat but still, IMHO, is safer to remove them after LM1117s.
I just connect my Headphone amp WOO6 .....and listen to the music .
Did you take in account the peculiar LPF needed by AK4396 (fc=125kHz, Q=0.692)?
Can you explain me what does it mean ?
Thanks
Serge
Hi Serge,
those AKM DACs don't do the entire LP filtering on chip and they depend from the external LPF to give a straight frequency response.
AK4393/5 and AK4396 have a different external LPF (see datasheets) thus they give different frequency response which should be filtered accordingly.
In AK4396 datasheet is stated that LPF should have that cut-off frequency and that Q to have a balanced frequency response.
Since they are noise shaping DAC's there will be a lot of noise produced above the audio band requiring the use of a low pass filter to attenuate it.The AK4396 is much better than the AK4393 in this regard so the low pass filters corner frequency can be shifted higher.
Thanks Claymen, i didn't realise I had the 4396 version. I will change the caps to the recommended and let you know the results. Fingers crossed
yes Dario i'm aware of that since I used the LPF in the datasheet on my 4396 board.My point ,though not clear was that the 4396 is claimed to have less hi frequency noise (similar to an R2R dac)than typical for a Delta-sigma dac.Therefore AKM suggest the use of a filter at a higher frequency than usual 125k .
Maybe my English is not so good but I'm not sure that I understand the quoted sentence.
BTW the AK4396 LPF (fc =125kHz) do more than filtering HF noise (quite non-existent with AK4396), it also balance the frequency response.
Earlier in this thread I've posted two graphs that shows how different is the effect of AK4393 and AK4396 LPFs
Batterian PSU
the 5V rail recommendation from AKM data sheet is 4.75 - 5.25V. Max 6V ok?
the 3.3V rail between 3V - 5.25V max 6V ok?
it means 4 of AA batteries 6V can handle it, doesn't? 4+4 pack for more lasting.
the 3.3V rail for CS8416 recommendation from CirrusLogic is 3.1 - 3.46V and again max 6V!
and again, a pack of 3 AA batteries 4.5V would not bee nice? or 3+3 more lasting. quiet cheap.
appreciate for any . . .
the 5V rail recommendation from AKM data sheet is 4.75 - 5.25V. Max 6V ok?
the 3.3V rail between 3V - 5.25V max 6V ok?
it means 4 of AA batteries 6V can handle it, doesn't? 4+4 pack for more lasting.
the 3.3V rail for CS8416 recommendation from CirrusLogic is 3.1 - 3.46V and again max 6V!
and again, a pack of 3 AA batteries 4.5V would not bee nice? or 3+3 more lasting. quiet cheap.
appreciate for any . . .
OT post moved to Shiga Clone thread per member request.
The exact value is 2.4K but I can't find PRPs in that value, so I've buyed 2.2K.
They works perfectly and the tonal balance seems right.
So if you can find 2.4K better, if not 2.2K are also OK.
Heyyas,
in the last days i've invested quite some time to rework my board. I've applied excessive amounts of flux and made sure every solder joint is perfectly clean and not "dry". I've then scratcher (with a small screwdriver) between *all* the close pads/joints. After that i soaked the board with some "cleaning alcohol" (don't know whats the english word for it, sorry) and scrubbed it clean with q-tips. Very very clean - i've done this multiple times until i could see no remains between any closer pair of pins. Everything that remains is some of the (again, missing a word...) "barfing agent" which is in the cleaning-alcohol that doesn't evaporate - a slight grey fog. Not trusting this i've measured the resistance between some very close pins with much of that agent, its quite too high for my DMM to measure.
I've also cleaned the board with a vacuum cleaner and blown every small particles off with a pressure-air-gun. Then i've checked the damn thing with the schematic another couple of times. On the other board i have all i've done is reversing the cap close to the opa into the right direction (both of them still work, as i've tested off-board).
Then i've tested both boards. Both still do get "hot", as in ~50-55*C for the 7812, ~50*C for the 1777-5v and ~45*C for the AK. Tough luck - it still happens on both boards. On the other side, as it *does* happen on both boards i doubt this is a bad-joint-problem. As it only happens to the ak/1117-5v/7812, not to the other 1117s, or 7912/7805 i doubt its a oscillation or otherwise regulator/capacitor related problem. As it does happen on both boards, a bad AK is unlikely.
So ... here i'm standing, still as dum as before :/.
Thanks again and kind regards
- Dario
PS: as everything works perfectly, i'm starting to thinking that i might leave it like this. On the other hand - i really really don't see why the f... my DAC should be hotter than my gainclone ... *grrr*
in the last days i've invested quite some time to rework my board. I've applied excessive amounts of flux and made sure every solder joint is perfectly clean and not "dry". I've then scratcher (with a small screwdriver) between *all* the close pads/joints. After that i soaked the board with some "cleaning alcohol" (don't know whats the english word for it, sorry) and scrubbed it clean with q-tips. Very very clean - i've done this multiple times until i could see no remains between any closer pair of pins. Everything that remains is some of the (again, missing a word...) "barfing agent" which is in the cleaning-alcohol that doesn't evaporate - a slight grey fog. Not trusting this i've measured the resistance between some very close pins with much of that agent, its quite too high for my DMM to measure.
I've also cleaned the board with a vacuum cleaner and blown every small particles off with a pressure-air-gun. Then i've checked the damn thing with the schematic another couple of times. On the other board i have all i've done is reversing the cap close to the opa into the right direction (both of them still work, as i've tested off-board).
Then i've tested both boards. Both still do get "hot", as in ~50-55*C for the 7812, ~50*C for the 1777-5v and ~45*C for the AK. Tough luck - it still happens on both boards. On the other side, as it *does* happen on both boards i doubt this is a bad-joint-problem. As it only happens to the ak/1117-5v/7812, not to the other 1117s, or 7912/7805 i doubt its a oscillation or otherwise regulator/capacitor related problem. As it does happen on both boards, a bad AK is unlikely.
So ... here i'm standing, still as dum as before :/.
Thanks again and kind regards
- Dario
PS: as everything works perfectly, i'm starting to thinking that i might leave it like this. On the other hand - i really really don't see why the f... my DAC should be hotter than my gainclone ... *grrr*
NebuK, I think the temperatures you measured are correct.
Consider the 1117-5: the dropout is 12-5 = 7 V, the current drawn from the AK4396 is 32 mA (from AK datasheet, typ. value), so the power dissipated in the 1117 is 224 mW.
The thermal resistance of the 1117 (junction to ambient) is 136°C/W, so the junction temperature will be 30.5°C over the ambient; assuming ambient is 22°C, the junction temperature will be 52.5°C.
The junction to case thermal resistance of the 1117 is 15°C/W, so the 1117 case will be 3.6°C less than the junction, that is about 49°C, very near to the one you measured (50°C)
You don’t have to be worried about these temperatures, they are quite normal.
Consider the 1117-5: the dropout is 12-5 = 7 V, the current drawn from the AK4396 is 32 mA (from AK datasheet, typ. value), so the power dissipated in the 1117 is 224 mW.
The thermal resistance of the 1117 (junction to ambient) is 136°C/W, so the junction temperature will be 30.5°C over the ambient; assuming ambient is 22°C, the junction temperature will be 52.5°C.
The junction to case thermal resistance of the 1117 is 15°C/W, so the 1117 case will be 3.6°C less than the junction, that is about 49°C, very near to the one you measured (50°C)
You don’t have to be worried about these temperatures, they are quite normal.
Was looking at the full BOM posted by ClaveFremen (Dario) in post 1125:
http://www.diyaudio.com/forums/digi...kit-cs8416-ak4393-5532-a-113.html#post2876581
Only two questions on it. Can I replace the 0.1uF Evox-Rifa SMR cap with either Wima pulse film or poly film? Its out of stock at Mouser and couldn't find it at Digikey and Newark. Also, does anyone have a part number or name for the two inductors are (L1 and L2)? Kudos to Clave and all the others for all the testing that was done with this DAC. Was looking at the CS8416 datasheet and are you sure that L1 and L2 are not ferrite beads?
Thanks,
Philippe
http://www.diyaudio.com/forums/digi...kit-cs8416-ak4393-5532-a-113.html#post2876581
Only two questions on it. Can I replace the 0.1uF Evox-Rifa SMR cap with either Wima pulse film or poly film? Its out of stock at Mouser and couldn't find it at Digikey and Newark. Also, does anyone have a part number or name for the two inductors are (L1 and L2)? Kudos to Clave and all the others for all the testing that was done with this DAC. Was looking at the CS8416 datasheet and are you sure that L1 and L2 are not ferrite beads?
Thanks,
Philippe
Last edited:
Looked at some more photos of the board and I think L1 and L2 are ferrite beads. I put the Panasonic ones from Mouser (667-EXC-ELSA35) on my copy of the BOM. Also am substituting Vishay/Roderstein metallized polypro cap from Partsconnexion (VISROD-72943) for the Evox-Rifa one. Let me know if I'm off the beam on them.
Thanks,
Philippe
Thanks,
Philippe