General > General Technical Chat

What ever happened to TV technicians?

<< < (12/17) > >>

Zoli:

--- Quote from: magic on May 09, 2022, 12:54:09 pm ---...
I see a bunch of jellybeans and those weird linear IGBTs from Toshiba.
Do you think the latter are irreplaceable in this design or that they don't make 2N5551 like they used to?
...

--- End quote ---
Either my google-fu fails me, but I cannot find any P-channel IGBT's available to buy; beside, what do you recommend to replace the MPS-Ux0's(originals, or suggested replacements are unobtanium)?
The only way I can think to replace the IGBT+MPS-U's are darlingtons; the problem with those are the fT, typically around 60MHz, which gives you a bandwith around 500kHz; the rest of the parts are available as originals or improved versions, AFAIK.

--- Quote from: magic on May 09, 2022, 12:54:09 pm ---...
BTW, Cordell's error-corrected MOSFET amplifier has been characterized in similar manner and beats this thing in most every way except DIM, apparently. That's 100% jellybean discretes, vertical MOSFET output stage.

--- End quote ---
Link, please; with proper parts list, so we can discuss what is jellybean and what unobtanium.

magic:
Hmm, you've got me, there is no complete parts list in the original article. It can be found on Cordell's website:
http://www.cordellaudio.com/papers/

At a quick glance, whatever JFETs he used are likely obsolete and the MOSFETs are IRF parts in TO-3 and the N-ch is obsolete so it would need to be replaced or the whole thing redesigned with newer plastic parts. I suppose you could just go to DIYAudio and ask him there if he (or anyone else for that matter) thinks an updated version would be feasible today. Maybe something has actually been built, I honestly have no clue.

P-IGBTs aren't a thing anymore, not even in switching. I just fail to see how they are supposed to be better than alternatives like ordinary power MOSFETs. (The author mentions cost as a factor). It is my understanding that IGBTs are fairly sluggish too; they are minority carrier devices, essentially a MOSFET/BJT Sziklai pair.

Zoli:

--- Quote from: magic on May 09, 2022, 09:42:36 pm ---Hmm, you've got me, there is no complete parts list in the original article. It can be found on Cordell's website:
http://www.cordellaudio.com/papers/

At a quick glance, whatever JFETs he used are likely obsolete and the MOSFETs are IRF parts in TO-3 and the N-ch is obsolete so it would need to be replaced or the whole thing redesigned with newer plastic parts. I suppose you could just go to DIYAudio and ask him there if he (or anyone else for that matter) thinks an updated version would be feasible today. Maybe something has actually been built, I honestly have no clue.

P-IGBTs aren't a thing anymore, not even in switching. I just fail to see how they are supposed to be better than alternatives like ordinary power MOSFETs. (The author mentions cost as a factor). It is my understanding that IGBTs are fairly sluggish too; they are minority carrier devices, essentially a MOSFET/BJT Sziklai pair.

--- End quote ---
IGBT vs. MOSFET: beside the price, the input capacity make the IGBT's favourite for the output stage; to have similar performances(SOA included), the MOSFET has multiple times the input capacity of the IGBT's: Ciss GT20D201-Y:1450pF, IXTH32P20T:14.5nF.
The start of the discussion was:

--- Quote from: NiHaoMike on May 08, 2022, 08:10:19 pm ---
--- Quote from: cdev on May 08, 2022, 04:56:17 pm ---Old audio equipment is often better than new audio equipment so is often worth fixing.

--- End quote ---
I assume you're referring to build quality since modern transistors and opamps perform vastly better than old ones. They're still improving albeit well into diminishing returns.

I also read claims that old music (especially from the days of tube amps) was mastered with technical limitations of that era consumer equipment in mind and sound "wrong" on modern equipment, similar to retro games looking really poor on a modern gaming monitor.

--- End quote ---
I think at this moment we can stop, and agree, that currently a wide-band, high powered linear amplifier is a real challenge to build, specifically caused by the lack of the proper output&driver stage.

Circlotron:

--- Quote from: Zoli on May 09, 2022, 11:23:55 pm ---IGBT vs. MOSFET: beside the price, the input capacity make the IGBT's favourite for the output stage; to have similar performances(SOA included),

--- End quote ---
From what I understand, IGBTs have a smaller die than a MOSFET of similar rating so I'm surprised that they have a comparable SOA. Does the "bipolar" part of the IGBT name indicate that they would be subject to second breakdown like a traditional bipolar transistor?

Zoli:

--- Quote from: Circlotron on May 10, 2022, 12:05:09 am ---
--- Quote from: Zoli on May 09, 2022, 11:23:55 pm ---IGBT vs. MOSFET: beside the price, the input capacity make the IGBT's favourite for the output stage; to have similar performances(SOA included),

--- End quote ---
From what I understand, IGBTs have a smaller die than a MOSFET of similar rating so I'm surprised that they have a comparable SOA. Does the "bipolar" part of the IGBT name indicate that they would be subject to second breakdown like a traditional bipolar transistor?

--- End quote ---
Had - past tense, since the linear IGBT's became unobtanium; and I think I saw the second breakdown line on the the linear IGBT's SOA graphic.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod