LM317 re configuration to 20V..
Hi
after checking my boxes i changed my mind and I desolder the 4300 ohms and used a normal resistor 1/4W- 3k6 (3584) to set the LM317 to 20V.
i used a 47µF for output and a 220n (blue) for input of the LM317. both caps are missing at the pcb.
i checked the Voltages:
IN 23,97V
adj 18,77V
OUT 20,03V
so the original psu can be used correctly.
the 2nd pic you can see the parallel wima caps 150nF/63V to the 1800µF Rubycon -back side.
test will be done..................
chris
Hi
after checking my boxes i changed my mind and I desolder the 4300 ohms and used a normal resistor 1/4W- 3k6 (3584) to set the LM317 to 20V.
i used a 47µF for output and a 220n (blue) for input of the LM317. both caps are missing at the pcb.
i checked the Voltages:
IN 23,97V
adj 18,77V
OUT 20,03V
so the original psu can be used correctly.the 2nd pic you can see the parallel wima caps 150nF/63V to the 1800µF Rubycon -back side.
test will be done..................
chris
Attachments
Last edited:
Attachments
Yep. In measurement pictures here, by accident also one of lm317 without output capacitor (outputcap can cause peak so one of options they also tried was capless).
Simple Voltage Regulators Part 1: Noise - [English]
Simple Voltage Regulators Part 1: Noise - [English]
Has anyone figured out what this IRF9530 is used for? I'm getting curious and I only have a TPA3255 without such a P-channel power-FET.
no idea...its getting very hot if you push 2vrms into the RCA inputs !!!
see some posts before....
??
2vrms input would be like maximum output if opamps are used as buffer only, could be way above maximum, so hard clipping, when opamps are not just buffers.
Seems the mosfet is less switching off when psu inputvoltage is higher, maybe amp was intended for higher voltage, the setting of lm317 seem to suggest that too.
Seems the mosfet is less switching off when psu inputvoltage is higher, maybe amp was intended for higher voltage, the setting of lm317 seem to suggest that too.
The mosfet only switches power to the pvdd pins tpa3250 and lm317. The part you pictured shows components near dc entry setting the gate to source voltage but also show some components at side of pcb going towards other side of tpachip (front), probably mcu, I don't know, pictures do not show. So gatevoltage isn't just set by components at dc entry. Components at dc entry seem just to make mosfet switch on when polarity is correct, but maybe they need to keep voltage close to mosfet switching off to be able to have effect from the parts going to other side(front). I think the mosfet starts switching on off at a very high frequency, then it gets hot. The tpachip and lm317 have the rubycons reservoir to bridge the off time but maybe with heavy long lasting bass at time mosfet is off somehow, let voltage drop enough for own tpaprotection to trip undervoltage protection.
sound check both amps LM49720
Hello
Both amps are with the LM49720 op amps now.
amp 1 is the regulated 20V (LM317) version with the caps as you can find in posts before.
amp 2 is the original, just the op amps
the psu is from amp 2 and its powering with 24V...so it is my target for now.
sound check:
in every track the amp 1 is better
in case of:
better timing, clearer details, better dynamic = in same tracks the voices and instruments are clear defined in pause and play.
the darkness is better and in the listening centrum more room is available for the singer and instruments...the amps is playe more "free"
so this 8 Euro (opamps 3 each, caps and resitor) bucks tuning is working fine !!!thanks for your help Fauxfrench, Irribeo and others.. ! 
chris
next .....: any more modding ideas...?
changing the grey output caps?
Hello
Both amps are with the LM49720 op amps now.
amp 1 is the regulated 20V (LM317) version with the caps as you can find in posts before.
amp 2 is the original, just the op amps
the psu is from amp 2 and its powering with 24V...so it is my target for now.
sound check:
in every track the amp 1 is better
in case of:
better timing, clearer details, better dynamic = in same tracks the voices and instruments are clear defined in pause and play.
the darkness is better and in the listening centrum more room is available for the singer and instruments...the amps is playe more "free"
so this 8 Euro (opamps 3 each, caps and resitor) bucks tuning is working fine !!!
thanks for your help Fauxfrench, Irribeo and others.. ! chris
next .....: any more modding ideas...?
changing the grey output caps?
How is switch off behavior at higher music volumelevels now ?
Your speakers are easy load for normal class AB amps, but well above impedance the normal tpa3250 outputfilter is designed for. With other inductors this gives easily audible added sound, I think this might be a little less coloured with the aircores here, only that behavior, maybe, I have no idea how aircores influence each other and everything else in amp and room.
Can you measure the inductance of the aircores ?
Your speakers are easy load for normal class AB amps, but well above impedance the normal tpa3250 outputfilter is designed for. With other inductors this gives easily audible added sound, I think this might be a little less coloured with the aircores here, only that behavior, maybe, I have no idea how aircores influence each other and everything else in amp and room.
Can you measure the inductance of the aircores ?
measurement of the coils.....
Hi irribeo
hopefully the blue (v3) and the white (v5) are the same. i measured the free amp. = no modding , the origianl one white pcb ...
tool = LCR Peakteck 2170
10,5µH
10,7µH
11,2µH
10,9µH
thx
chris
Hi irribeo
hopefully the blue (v3) and the white (v5) are the same. i measured the free amp. = no modding , the origianl one white pcb ...
tool = LCR Peakteck 2170
10,5µH
10,7µH
11,2µH
10,9µH
thx
chris
morning....
i cant follow you. because in the datasheet- TPA3250 - there is written 10µH for 80hms too.
please can you explain which datas you set into the the e.g. TI calculator?
LCFILTER-CALC-TOOL Class-D LC Filter Designer | TI.com
sonus faber venere 1.5. = more than 8 ohm ....
Sonus Faber Venere 1.5 loudspeaker Measurements | Stereophile.com
yes, that graph I looked at to come to around 20 ohm at filterfrequency, meaning the venere will have a little boosted treble from 2KHz and progressively up, also the LC filter will ring a little, impulsrespons.
depending on capvalue on board the filter is critically damped for 4 or 6 ohm, overdamped for anything below that, underdamped anything above that, the boosting/peaking/ringing I meant.
depending on capvalue on board the filter is critically damped for 4 or 6 ohm, overdamped for anything below that, underdamped anything above that, the boosting/peaking/ringing I meant.
yes i know the sonus are not the best. i will change the filter in the future---because its "sounded" for 5.1 i guess......
...., but actually its not bad. i will switch to my KEf Q100...and report the sound difference with the 2 amps
its the amp we are talking about, not the speaker
the amp doesn't have a flat frequency respons with the sonus faber load and the amp will add sparkle and mist with the sonus faber as load
the kef is a different load, the amp!!!!! will have a flatter respons with the kef load and will add less sparkle and mist.
on your scope you'll see less transient ringing and less rising frequency respons when you test your amp with a load like the kef. the kef can have unbearable piercing treble and the sonus faber can sound mellow sweet, I don't know, I only focus on the amplifier part
the amp doesn't have a flat frequency respons with the sonus faber load and the amp will add sparkle and mist with the sonus faber as load
the kef is a different load, the amp!!!!! will have a flatter respons with the kef load and will add less sparkle and mist.
on your scope you'll see less transient ringing and less rising frequency respons when you test your amp with a load like the kef. the kef can have unbearable piercing treble and the sonus faber can sound mellow sweet, I don't know, I only focus on the amplifier part