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

[thimios]

Hi Thimios, thank you for the spectrums - looking good.

I think overall distortion level may be even lower if you roughly double the bias - as I remember, my optimum value was 62.5mA of idle current per output pair.
It's interesting that distortion level is rather consistent throughout the bandwidth.

Cheers,
Valery

Hi Valery,idle is already 31mA/piece=62mA/pair.
I have something interesting too.
Another test using a real speaker as load.

 

[thimios]

AX-Deluxe FFT Test.

Using a real 3way speaker as load.

 

[pinnocchio]

This is really exciting! I might actually be able to setup my first hybrid amplifier!

Do

 

[Kay Pirinha]

Are there TubSuMo IPS PCB's readily available, either in through-hole or in SMD version?


Best regards!

 

[jwilhelm]

I still have three of the original TubSuMo input boards left, plus a bunch of the output boards. I have some of the SMT version for the NS series output boards, but would like to test one to verify operation before making them available.

 

[vzaichenko]

Hi Valery,idle is already 31mA/piece=62mA/pair.
I have something interesting too.
Another test using a real speaker as load.

Emmm... in the complementary pair, transistors (as well as emitter resistors) are connected in series. Meaning that the voltage drop across those resistors sums up, but the current stays the same.

So, you need to have 14mV across each 0.22R resistor, or 28mV across 2 resistors in series

 

[thimios]

Emmm... in the complementary pair, transistors (as well as emitter resistors) are connected in series. Meaning that the voltage drop across those resistors sums up, but the current stays the same.

So, you need to have 14mV across each 0.22R resistor, or 28mV across 2 resistors in series

What a stupid fault!
Sorry,i will test again.

 

[Vrystaat]

Only a little footfault , no harm done !!

 

[thimios]

AX-Deluxe New Bias

17mA/0.22R Looks like the optimum.

 

[thimios]

AX-Deluxe+NS

17mV/0.22R
On a real speaker.

 

[vzaichenko]

Harmonics 2 and 3 with dominant 3, but rather low absolute values and rather clean high-order areas. I like the way it looks. The amp is also rather tolerant of bias variations - previous measurements with less than 50% bias setting are still good enough

Did you audition the thing?
It will be interesting what TubSuMo front-end is going to show under the same conditions.

 

[jwilhelm]

Distortion numbers look better with a speaker attached. I though it would be the other way around.

 

[pinnocchio]

My build is coming along nicely. That is the NS OPS with Vertical CFA IPS

Here's the regulator PCB





Here's the Vertical CFA





I'm pretty happy so far. I'm also working on another project at the same time but it is a headphone amp.

Ciao!
Do

 

[pinnocchio]

Hi Jeff,

Could you reserve me a set of Tubsumo IPS for the NS OPS?

Thanks
Do

 

[thimios]

Harmonics 2 and 3 with dominant 3, but rather low absolute values and rather clean high-order areas. I like the way it looks. The amp is also rather tolerant of bias variations - previous measurements with less than 50% bias setting are still good enough

Did you audition the thing?
It will be interesting what TubSuMo front-end is going to show under the same conditions.

Just for a little but not on main speakers(i haven't protection board yet).
I think that is a nice sounding IPS.
When i will have the TubSumo on hand we will know the difference.

Distortion numbers look better with a speaker attached. I though it would be the other way around.

That's why we run the test!
Real world is different than simulations.

 

[thimios]

My build is coming along nicely. That is the NS OPS with Vertical CFA IPS

Here's the regulator PCB

View attachment 647350

View attachment 647351

Here's the Vertical CFA

View attachment 647352

View attachment 647353

I'm pretty happy so far. I'm also working on another project at the same time but it is a headphone amp.

Ciao!
Do

Nice built!
Probably is a new version of the power supply.

 

[thimios]

AX-Deluxe using MAT02

Today tried the MAT02 as input pair but without success.
I see a big distortion level at low volume due to oscillation?
hfe=860,probably the input stage current is realy big and must be lowered?

 

[AndrewT]

Today tried the MAT02 as input pair but without success.
I see a big distortion level at low volume due to oscillation?
hfe=860,probably the input stage current is realy big and must be lowered?

or increase the input emitter degeneration,
or do both, i.e. increase degen and reduce Ic.

 

[vzaichenko]

Hi Thimios,

Are you sure the pinout is correct (I mean, the CBE leads are connected as expected)?
Input stage current is around 1.3mA per collector and it does not depend on hFE of the pair.
By the way, hFE of 2N5089 you used initially is normally around 1000.

So, my guess is the wrong pinout

Cheers,
Valery

 

[thimios]

Hi Thimios,

Are you sure the pinout is correct (I mean, the CBE leads are connected as expected)?
Input stage current is around 1.3mA per collector and it does not depend on hFE of the pair.
By the way, hFE of 2N5089 you used initially is normally around 1000.

So, my guess is the wrong pinout

Cheers,
Valery

Because i haven't any 2N5089 i used 2N5088 in my board.
Yes i'm sure for pinout,pins c,e reverced in one side.