Revisiting some "old" ideas from 1970's - IPS, OPS

[thimios]

Just a quick listening test would tell if the noise is still there. Mine had a quite noticeable 60hZ hum when the input was disconnected, but went away as soon as anything was hooked to it.

Ok Jeff i will come back soon
Also i will try to lift zeners and separate the VAS & INP power supply.

 

[thimios]

Just a quick listening test would tell if the noise is still there. Mine had a quite noticeable 60hZ hum when the input was disconnected, but went away as soon as anything was hooked to it.

Yes the noice with an empty input is still here.
This test will be more useful.
Two softwares for sure.

 

[thimios]

NS FRANKENSTEIN!

Here is another Frankenstein,the Ostripper Symasui Survivor IPS +NS OPS.
This looks like a good combination.
The only bad thing is the long time(30'')before offset settle from 3v d.c to zero.

 

[AndrewT]

the 2u2F & 680k have a very low roll-off. You could try 1uF and 470nF to see how quickly they settle and how effectively they remove offset.

The 22k feedback of the servo is quite low. Does your version show 4V as the U1 output? That 4V indicates an enormous current passing through R2 & R24
What do you measure as the voltages at these resistors?

 

[jwilhelm]

Is this the version with the updated values to speed up settling time, or the original?

 

[thimios]

the 2u2F & 680k have a very low roll-off. You could try 1uF and 470nF to see how quickly they settle and how effectively they remove offset.
The 22k feedback of the servo is quite low. Does your version show 4V as the U1 output? That 4V indicates an enormous current passing through R2 & R24
What do you measure as the voltages at these resistors?

Is this the version with the updated values to speed up settling time, or the original?

Andrew i will try to see what you want to know.
Jeff,i don't know what is the correction you speaking about.
Can you give me some details?

 

[jwilhelm]

I'm trying to find the schematic. We noticed the slow settling time was tripping my protection, so Pete revised it for faster settling.

 

[jwilhelm]

Here's one updated schematic. There's a newer one, but I haven't found it yet.

 

[evanc]

Quote from OS:
I see there is discussion about servo values. I have standardized with
a simple combo of 1 meg/1uF for all my servo'ed IPS's.
As I mentioned before , 5.6K for the spook/symasui servo outputs.
(470R for ND/C servo's)

 

[thimios]

Here's one updated schematic. There's a newer one, but I haven't found it yet.

Quote from OS:
I see there is discussion about servo values. I have standardized with
a simple combo of 1 meg/1uF for all my servo'ed IPS's.
As I mentioned before , 5.6K for the spook/symasui servo outputs.
(470R for ND/C servo's)

Thanks Jeff i will try to find what are the values that was used.
Evan,these 5k6,470R are in position of R28=22K?

 

[thimios]

Using Spectraplus.

 

[vzaichenko]

Distortion is low and good profile

 

[thimios]

Distortion is low and good profile

Yes Valery,that's my opinion too!
May be it's time for an smd version to match the NS!
This will be a job for you Jeff

 

[xrk971]

Thimios,
Really nice measurements. Which software package do you like best for making the measurements? You must have tried 4 different ones?

 

[jwilhelm]

Yes Valery,that's my opinion too!
May be it's time for an smd version to match the NS!
This will be a job for you Jeff

I've been thinking of doing a version of the Slewmasters to match the same input board size. I'll see what Ostripper thinks of the idea.

 

[thimios]

Thimios,
Really nice measurements. Which software package do you like best for making the measurements? You must have tried 4 different ones?

Try all to see what is more suitable for you.
RMAA is all in one solution.
ARTA is good but a little analytic in the noice figure area.
Spectraplus isn't free.I have a trial(30days)version and i used this 2-3 days ago,looks good up to now.
I wish i have a better sound card but isn't possible.

 

[thimios]

I've been thinking of doing a version of the Slewmasters to match the same input board size. I'll see what Ostripper thinks of the idea.

I'm so glad seeing Pete here again

 

[thimios]

Trying to make the offset settle time shorter i see that decreasing the R28 value has as result a higher distortion .
My first try with 5k6 in R28 position(servo out).
Then with R28=10k.
Last changeR28= 15k.

 

[vzaichenko]

I would leave R28 as is (22K), but decrease C12, C13 to 1uF.
It will roughly half the settling time, but still leave the servo's time constant high enough.

 

[thimios]

I would leave R28 as is (22K), but decrease C12, C13 to 1uF.
It will roughly half the settling time, but still leave the servo's time constant high enough.

Ah sorry my friend, i forgot to say that these are 1uf from the beginning