eBay mono LM1875 kit

[chermann]

Did all the swapping R to L etc all the way through. Not a problem as it's pulled apart and have moved on.

...sorry
ask ESP directly?


chris

 

[rabbitz]

Not worth the effort as it was never going to be used as I have too many not being used. I've built over 30 SS amps and this is the only one that beat me.

 

[chermann]

Not worth the effort as it was never going to be used as I have too many not being used. I've built over 30 SS amps and this is the only one that beat me.

30 : 1 is a very good quote !!

 

[chermann]

...measurments with caps at the speakers terminal

Hi


i was motivated to try a square wave test at my listening "mock up amp".
i use a 220nF at one channel to look at . the cap is a grey Kemet 220nF MKP X2 / 275VAC load is 8R non inductive on both channel.
its about 10,8 Watt each channel


pic 1 200mVrms in square 10khz_220nF
pic 2 200mVrms in square 20khz_220nF
pic 3 400mVrms in square 10khz_220nF
pic 4 400mVrms in square 20khz_220nF

oscillating..
pic 5 700mVrms in square 10khz_220nF
pic 6 700mVrms in square 20khz_220nF


after this bad oscillating i will not do 4R test with max power...


some comments or ideas?
chris

 

[chermann]

its gooing worse with 440nF

here is the same test but i try the other channel - L channel but with 440nF...same cap type.


now with just 200mVrms at 8R its about 2,6Watt per channel



pic1 200mVrms in square 1khz_440nF_L channel
pic2 200mVrms in square 1khz_440nF_L channel_2
pic3 200mVrms in square 10khz_440nF_L channel
pic4 200mVrms in square 20khz_440nF_L channel
pic5 200mVrms in square 30khz_440nF_L channel

 

[chermann]

...bad at 440nF and 400mVrms input

pic1 400mVrms in square 1khz_440nF_L channel
pic2 400mVrms in square 1khz_440nF_L channel_2
pic3 400mVrms in square 10khz_440nF_L channel
pic4 400mVrms in square 20khz_440nF_L channel
pic5 400mVrms in square 30khz_440nF_L channel


its about 10,5 Watt at 8R

 

[chermann]

As i wrote i listening since weeks and i found to use a "zobel 220nF +10R" at the speakers output as better sound.
the only way i can explain "technically" is a comparison. no zobel is the swing bigger = overshoot and it is a longer swing



pic1 400mVrms in square 10khz_440nF_L channel
pic2 400mVrms in square 10khz_440nF_L channel - no zobel
pic3 400mVrms in square 20khz_440nF_L channel
pic4 400mVrms in square 20khz_440nF_L channel - no zobel
pic5 400mVrms in square 30khz_440nF_L channel
pic6 400mVrms in square 30khz_440nF_L channel - no zobel


chris

 

[chermann]

Thanks to John audio tech for that idea = test the stabilisation of an amp
actually i am afraid to test with 4R load..

 

[chermann]

I ´d play around with max power test - i use a 100hZ tone.
here some compare data´s sinus vs square wave
no caps on the speakers just the zobel and the non inductive 8R.

sinus max input is 600mVrms = 40,4Vpp and 13,8Vrms at 8R its about 24Watt
the rails are in idle at my amp +/- 26,6V = 53,2V
during this max test i get a non sync. rail voltage drop with 23V and 23,3V = 46,3 i loose 6,9Volts.

square test is harder - but i was able to "climb" to the rail nearer...
max input is 900mVrms square = 42,5Vpp and 19,6Vrms at 8R its about 48Watt
the rails are in idle at my amp +/- 26,6V = 53,2V
during this max test i get a non sync. rail voltage drop with 21,7V and 22,1V = 43,8 i loose 9,4Volts.

chris
good night

 

[chermann]

...avoi oscillation..method 1 . add 47pF at feedback

Hi

i am not sure if this is the best method but i solder a 47pF cap in parallel to the 22k Feedback resistor. any other hints are very welcome.
R channel input is 400mVrms into 8R and the 440nF cap on the speaker terminal.

pic 1 400mVrms in square 10khz_440nF_R channel
pic 2 400mVrms in square 10khz_440nF_R channel + 47pF feedback
pic 3 400mVrms in square 20khz_440nF_R channel
pic 4 400mVrms in square 20khz_440nF_R channel + 47pF feedback
pic 5 400mVrms in square 30khz_440nF_R channel
pic 6 400mVrms in square 30khz_440nF_R channel + 47pF feedback


any comments/ideas ?
chris

 

[chermann]

recovered chips cannot do the same performance?

Hi


i want to try this square wave test on the other boards too. this boards are with the 1875 recovered - shortcut survived. the test is with my lab psu set to 27V, 3A each rail. 250mV is the maximum because after some seconds a kind of protection mode kicks in! every 2,5 seconds the chip is switching on/off.

pic 1 this is the board i tested.
pic 2 250mVrms square in 10kHz at 8R + 440nF cap
pic 3 250mVrms square in 20kHz at 8R + 440nF cap
pic 4 250mVrms square in 30kHz at 8R + 440nF cap
pic 5 show you with a long time set that every 2,5 seconds the chip is switching off -my psu shows me that the voltages drops to zero.
this phenomenon i get just at the other amp with max power and high frequency. so for me this chip actually are not 100% "recovered"
do somebody have the same experience??
chris

 

[FauxFrench]

Comment: Reducing bandwidth (47pF) seems to improve stability margin.
Idea: Try with 220pF instead of the 47pF.

 

[chermann]

Comment: Reducing bandwidth (47pF) seems to improve stability margin.
Idea: Try with 220pF instead of the 47pF.

Hello FF


Thank you for your advice ! ...i will do so.
chris

 

[chermann]

..220pf at the feedback

Hi


here is the result at my mock up amp. i did the test with 400mvrms into 8R and 440nF cap at the R channels output. L channel just 8R.

400mVrms in square 10khz_440nF_R channel_220pF feedback
400mVrms in square 20khz_440nF_R channel_220pF feedback
400mVrms in square 30khz_440nF_R channel_220pF feedback

chris

 

[FauxFrench]

Hi


here is the result at my mock up amp. i did the test with 400mvrms into 8R and 440nF cap at the R channels output. L channel just 8R.

400mVrms in square 10khz_440nF_R channel_220pF feedback
400mVrms in square 20khz_440nF_R channel_220pF feedback
400mVrms in square 30khz_440nF_R channel_220pF feedback

chris

Many thanks.
Too slow, 100pF should be right. Eventually, try to hear if you notice the difference between no capacitor and 100pF with music.

 

[chermann]

Merci FF...i will do so...

 

[chermann]

Hi FF


Actually i listening with 220pf at the feedback about 2.5 hours and compare to my TPA3255.
the lm 1875 is now a different - better amp - its clear and deeper sound stage, partly i go through some older album and its sounds very very nice.
nevertheless i will modify the cap to the 100pf to get faster response...


thanks

chris

 

[chermann]

...100pf in the feedback

Good morning


here is the test with my mock up amp (amp 1_2) with the feedback cap of 100pf each channel. test is with 8R and the Rchannel is the 440nF cap load added.


pic 1 400mVrms in square 10khz_440nF_R channel 100pF feedback
pic 2 400mVrms in square 20khz_440nF_R channel 100pF feedback
pic 3 400mVrms in square 30khz_440nF_R channel 100pF feedback


i tried 600mVvrms in ..it is stable

pic 4 600mVrms in square 10khz_440nF_R channel 100pF feedback
pic 5 600mVrms in square 20khz_440nF_R channel 100pF feedback
pic 6 600mVrms in square 10khz_440nF_R channel 100pF feedback


listening not done...(kefQ100)
as i wrote the post before - the sound - with 220pf - was very dynamic and clear without "noise/fog " in the background. the details get much better without harshness or bad high snake t,z,s..


chris

 

[chermann]

Hi


just a question : i use a ceramic style cap for this feedback - now 110pF. Should i use a polyprepylen cap (wima..) because of the bad vibration effect at the ceramic?


chris

 

[chermann]

frequence response with 8R and 4R resistive load

Good morning


after all this changes i did a frequency response with resistive load 8R and 4R. input is 10mmVrms and that´s about 0,7W @8R and 1,3W @ 4R.
its a soft roll off of the gain of 27db nearly the same for 8R or 4R

17khz -0,3dB
23khz -0,5dB
31khz -1dB
47khz -2db
59khz -3db


chris