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

[hahfran]

I've got the good models of NJW3281/1302, that I use for simulations in such cases. They are rather close in the key parameters and both work well in the real prototype. Well, Sankens allow more power.

Are the NJW3281/1302 with internal diode for temperature sensing?
From viewpoint of latency, (heat transfer is by diffusion hence slow) they should provide a faster reading of the actual junction temp ( which matters)

 

[AndrewT]

NJL are the TT type with the integrated diode.

 

[vzaichenko]

OK, tested with 0.1uF and with 1.1uF - the ones I've got in hand right now

Just to clarify this point - previous measurements with capacitice load were taken right at the capacitor, behind the coil.

Here are the oscilloscope photos, taken right at the cap (1.1uF in parallel with 8 ohm load) and then the one, taken at the same time right at the amplifier output (before the coil 1.5-2uH). This is TubSuMo IPS + Non-switching OPS

 

[damon1983]

Just to clarify this point - previous measurements with capacitice load were taken right at the capacitor, behind the coil.

Here are the oscilloscope photos, taken right at the cap (1.1uF in parallel with 8 ohm load) and then the one, taken at the same time right at the amplifier output (before the coil 1.5-2uH). This is TubSuMo IPS + Non-switching OPS

0,3F and transformer 2*30VAC*8,34A will be enough for this amp?

 

[vzaichenko]

0,3F and transformer 2*30VAC*8,34A will be enough for this amp?

Ideally - 2*36VAC. With 2*30VAC you will need to change some resistor values, slightly increasing distortion - this is already on the edge. However - still possible.

As an option, you can build a different combination - non-switching OPS with VERTICAL front-end - here:
http://www.diyaudio.com/forums/soli...e-old-ideas-1970s-ips-ops-39.html#post4561721

This one is easy to adapt to the lower rails. Also a great performer (CFA).

 

[damon1983]

Ideally - 2*36VAC. With 2*30VAC you will need to change some resistor values, slightly increasing distortion - this is already on the edge. However - still possible.

As an option, you can build a different combination - non-switching OPS with VERTICAL front-end - here:
http://www.diyaudio.com/forums/soli...e-old-ideas-1970s-ips-ops-39.html#post4561721

This one is easy to adapt to the lower rails. Also a great performer (CFA).

OK. 0,3F will be enough, or bigger need?

 

[vzaichenko]

OK. 0,3F will be enough, or bigger need?

This is more than enough. In my current build I use 6 x 10000 uF per channel - this is already enough.

 

[vzaichenko]

VERTICAL IPS + NS-OPS

Tested and measured Vertical IPS (CFA) with non-switching OPS.
Spectrums are 50W @ 8 ohm, square waves are 40V p-p, clipping is close to the rails (100V p-p).
Measured slew rate is around 100V/uS. Not bad

I will publish minor updates to the Vertical soon.

Virtual Zero (distortion) amplifier

After some minor corrections in VERTICAL IPS and certain tune-up of NS-OPS, amazing results are observed. See the spectrums... I did not believe what I see initially, so I have checked everything a few times, measured with some other front-ends, then with this one again. It's what it is. Measured at 50W @ 8 ohm dummy load (20V RMS) with +/-52V rails.

Sounds delicious, at least in my environment. Jeff - you've got to try and see (hear) Terry, you were also somewhere close as I remember... Bimo - this is also a good candidate for your experiments

Very, very cool one

Cheers,
Valery

 

[jwilhelm]

My new TubSuMo boards have arrived. Now all I need is some time to assemble everything.

 

[damon1983]

Virtual Zero (distortion) amplifier

After some minor corrections in VERTICAL IPS and certain tune-up of NS-OPS, amazing results are observed. See the spectrums... I did not believe what I see initially, so I have checked everything a few times, measured with some other front-ends, then with this one again. It's what it is. Measured at 50W @ 8 ohm dummy load (20V RMS) with +/-52V rails.

Sounds delicious, at least in my environment. Jeff - you've got to try and see (hear) Terry, you were also somewhere close as I remember... Bimo - this is also a good candidate for your experiments

Very, very cool one

Cheers,
Valery

congratulation

 

[vzaichenko]

congratulation

Thank you

 

[bimo]

Virtual Zero (distortion) amplifier

After some minor corrections in VERTICAL IPS and certain tune-up of NS-OPS, amazing results are observed. See the spectrums... I did not believe what I see initially, so I have checked everything a few times, measured with some other front-ends, then with this one again. It's what it is. Measured at 50W @ 8 ohm dummy load (20V RMS) with +/-52V rails.

Sounds delicious, at least in my environment. Jeff - you've got to try and see (hear) Terry, you were also somewhere close as I remember... Bimo - this is also a good candidate for your experiments

Very, very cool one

Cheers,
Valery

very

I will try to make a simulation to know how it work...
Can you make pdf file for this IPS?

I still do not understand clearly how your OPS work. Can you or someone compare this OPS with EF3 OPS with same IPS?

 

[Cortez]

Congrats Valery, a really great job!
I would be intrested in a small descreption on both IPS/OPS as well if possible.
Do you notice something special in the sound? Or is it "just" good and that's all?

 

[vzaichenko]

Congrats Valery, a really great job!
I would be intrested in a small descreption on both IPS/OPS as well if possible.
Do you notice something special in the sound? Or is it "just" good and that's all?

Hi Cortez,

I'm planning to write some more info, but right now please see some details on my web site (link in my signature below), also this post with setup guidelines may be useful:
http://www.diyaudio.com/forums/soli...e-old-ideas-1970s-ips-ops-35.html#post4554894

Cheers,
Valery

P.S. With regards to the sound - a true A/B test is required to really see the differences with other designs. Right now - it's just among the best ones Some details - on my web site again

 

[still4given]

Virtual Zero (distortion) amplifier

After some minor corrections in VERTICAL IPS and certain tune-up of NS-OPS, amazing results are observed. See the spectrums... I did not believe what I see initially, so I have checked everything a few times, measured with some other front-ends, then with this one again. It's what it is. Measured at 50W @ 8 ohm dummy load (20V RMS) with +/-52V rails.

Sounds delicious, at least in my environment. Jeff - you've got to try and see (hear) Terry, you were also somewhere close as I remember... Bimo - this is also a good candidate for your experiments

Very, very cool one

Cheers,
Valery

Hi Valery,

I had it all mounted and ready to fire. While waiting for you to get a chance to set yours up I did some reading about the outputs. Newark was out of them when I got ready to order so I bought some from an ebay seller who swore they were genuine. I read you could check them by filing the top edge and exposing the double die. A single die meant they are fakes. Guess what mine had. I pulled them to send back but the shipping was half the cost so I just chalked it up to bad judgement. When I can get some real ones I 'll get it mounted again. Meanwhile, could you post the asbuilt for both vertical and the NS OPS?

Thanks, Terry

 

[jwilhelm]

Hi Valery,

I had it all mounted and ready to fire. While waiting for you to get a chance to set yours up I did some reading about the outputs. Newark was out of them when I got ready to order so I bought some from an ebay seller who swore they were genuine. I read you could check them by filing the top edge and exposing the double die. A single die meant they are fakes. Guess what mine had. I pulled them to send back but the shipping was half the cost so I just chalked it up to bad judgement. When I can get some real ones I 'll get it mounted again. Meanwhile, could you post the asbuilt for both vertical and the NS OPS?

Thanks, Terry

The Sankens are single die, just larger. There is a TO247 inside the MT-200 plastic. Digikey is a Sanken dealer too.

 

[jwilhelm]

Here's couple I torture/tested to death.

 

[vzaichenko]

Hi Valery,

I had it all mounted and ready to fire. While waiting for you to get a chance to set yours up I did some reading about the outputs. Newark was out of them when I got ready to order so I bought some from an ebay seller who swore they were genuine. I read you could check them by filing the top edge and exposing the double die. A single die meant they are fakes. Guess what mine had. I pulled them to send back but the shipping was half the cost so I just chalked it up to bad judgement. When I can get some real ones I 'll get it mounted again. Meanwhile, could you post the asbuilt for both vertical and the NS OPS?

Thanks, Terry

Hi Terry,

Here are the schematics, exactly "as built":
http://www.diyaudio.com/forums/soli...e-old-ideas-1970s-ips-ops-41.html#post4578177

Hope you will find the good Sankens. If not - NJW3281/1302 will work as well, providing slightly lower current capability.

Cheers,
Valery

 

[evanc]

Valery,
Nice website.
As soon as I finish with the protection boards that Jeff has helped me with the NS OPS/Vertical IPS look like something I would like to try. Any chance of getting the manufacturing files. Would sanken 2sa1295/2sc3264 be suitable output devices...I have some on hand.
Thanks as always,
Evan


PS. I still have your TubSuMo in my system...best for me so far.

 

[Cortez]

I'm planning to write some more info, but right now please see some details on my web site (link in my signature below), also this post with setup guidelines may be useful:
http://www.diyaudio.com/forums/soli...e-old-ideas-1970s-ips-ops-35.html#post4554894

Thanks in advance! BTW: your site is very neat! (Just as your designs as I see so far..! )