peranders said:
Hi,
I don't wish to seem contradictory here, but such sweeping statements can be misleading.
These mosfets do come from the same family, and as such are nominally similar, but *usually* they have subtle differences in certain paramaters, such as RDs on,Ciss,Coss, etc. etc, and generally these associated capacitances increase with the higher Vds devices. (Much the same as most Toshibas & IRFs do too!)
I well know from experience that in many circuits these particular Hitachi mosfets can be a pig to stabilise properly (just ask Erno Borbely about this!), and the stabilisation requirements usually vary between the different types of mosfets because of the variations mentioned.
It does not follow that, although you can drop in a higher Vds device without giving consideration to this matter, any existing circuit stabilisation will still be satisfactory under all circumstances.
You have been warned, as the saying goes.
Regards,
Absolutely. Stability is very definitely the first thing to test when building a power amp using L-mosfets. Even if you use the same circuit and parts, your layout can make enough of a difference to affect the stability (depending, of course, on many other factors). In any case, I would assume that devices like these alternatives are similar enough to use without requiring any major changes in order to ensure the stability of the circuit. Just be sure to test first, and things should turn out OK.
peranders said:
Hi Peranders,
As I said, I didn't wish to cast doubts on what you had said, but having had a lot of experience with these Hitachi devices about 20 years ago, I though it best to mention the possibility.
Regrettably, I no longer have all of the specs for these devices to hand, so I could not readily check up on them, but I am surprised if *all* the specs are identical. This is particularly so, knowing how the different 'doping' of the substrates (usually needed for higher voltages) will also affect the 'associated' capacitances of the final chip.
I was not alone in discovering (hence the reference to Erno Borbely, who writes frequently in AudioXpress)
that swapping these Hitachi devices around, even though 'nominally' similar, could cause serious problems with stability, and several different 'versions' of stabilising arrangements were in vogue years ago.
However, if you say you know that *all* of the parameters (including the perhaps rather more 'obscure' ones I referred to)
are absolutely identical across these devices, I will accept what you say.
Certainly, several of the parameters *do* differ widely between the 'N' & 'P' counterparts, as I remember this very clearly, and will never forget it, as different compensations were needed (ideally) for the two different polarity devices.
Regards,
Lateral MOSFETs and stability
I've been playing with them for over 20 years also. I found out early on if you pay attention to layout, bypass appropriately/correctly for your application and keep the leads to the MOSFETs short, stability problems are nonexistent. I suppose this is common sense
Hopefully stating it anyway will help someone else.
I've built a number of Borbely projects. The only time I had to resort to his extra capacitors scheme was when I later found out I had a defective device (not failed, just WAY out of spec).
The "published" specs for the devices having different VDS max ratings are the same. It is reasonable to assume there is some "actual" difference due to processing that cause the devices to be binned in the different VDS range. In my experience, exchanging devices of different VDS max ratings caused no perceptible difference in performance, sound or stability. Yes, you may have to readjust bias, depending on your circuit.
In my opinion, these are not the best devices to make amps with, but they are VERY damage resistant (nice for experimenting) and the zero tempco bias point of ~100mA is convenient if you want a simple design.
mlloyd1
I've been playing with them for over 20 years also. I found out early on if you pay attention to layout, bypass appropriately/correctly for your application and keep the leads to the MOSFETs short, stability problems are nonexistent. I suppose this is common sense
Hopefully stating it anyway will help someone else.I've built a number of Borbely projects. The only time I had to resort to his extra capacitors scheme was when I later found out I had a defective device (not failed, just WAY out of spec).
The "published" specs for the devices having different VDS max ratings are the same. It is reasonable to assume there is some "actual" difference due to processing that cause the devices to be binned in the different VDS range. In my experience, exchanging devices of different VDS max ratings caused no perceptible difference in performance, sound or stability. Yes, you may have to readjust bias, depending on your circuit.
In my opinion, these are not the best devices to make amps with, but they are VERY damage resistant (nice for experimenting) and the zero tempco bias point of ~100mA is convenient if you want a simple design.
mlloyd1
Re: Lateral MOSFETs and stability
In my opinion, these are not the best devices to make amps with, but they are VERY damage resistant (nice for experimenting) and the zero tempco bias point of ~100mA is convenient if you want a simple design.
mlloyd1 [/B][/QUOTE]
Hi,
What you say is interesting, and it will be useful to several posters to know what you and Peranders have to say here on the full specs, and I agree with the comment that Hitachis are not (now) the best.
Whilst in all the time I played about with these various Hitachis, I never actually 'blew' one (!) so as you suggest they are very rugged, I did have a considerable problen on some occasions with oscillation in the Mhz regions.
Hence the (ultimate) use of asymmetrical compensation.
From the many latters in TAA at about that time, I was certainly not alone in this, it seemed!
I don't favour using high-value gate-stoppers in order to 'subdue' mosfets, but like yourself, it seems, using short leads and plenty of capacitance right on the pins, does help a lot with stability issues.
When I moved over to trying some Toshibas (plastic pacs with generally better specs, and certainly better sonic performance) again, like yourself, virtually all of my stability difficulties disappeared at a stroke, so whenever using mosfets now, I have stayed with Toshibas (and occasional IRFs).
Regards,
In my opinion, these are not the best devices to make amps with, but they are VERY damage resistant (nice for experimenting) and the zero tempco bias point of ~100mA is convenient if you want a simple design.
mlloyd1 [/B][/QUOTE]
Hi,
What you say is interesting, and it will be useful to several posters to know what you and Peranders have to say here on the full specs, and I agree with the comment that Hitachis are not (now) the best.
Whilst in all the time I played about with these various Hitachis, I never actually 'blew' one (!) so as you suggest they are very rugged, I did have a considerable problen on some occasions with oscillation in the Mhz regions.
Hence the (ultimate) use of asymmetrical compensation.
From the many latters in TAA at about that time, I was certainly not alone in this, it seemed!
I don't favour using high-value gate-stoppers in order to 'subdue' mosfets, but like yourself, it seems, using short leads and plenty of capacitance right on the pins, does help a lot with stability issues.
When I moved over to trying some Toshibas (plastic pacs with generally better specs, and certainly better sonic performance) again, like yourself, virtually all of my stability difficulties disappeared at a stroke, so whenever using mosfets now, I have stayed with Toshibas (and occasional IRFs).
Regards,
hexfet replacement
I found some Hexfets (irfp240 and Irfp9240) to replace the sj162/sk1058 (which obviously replace sj50/sk135). However I also read that they are not always suitable as replacement. They are a lot cheaper however (I can get them for around $2) and certainly worth a try. Does anybody have some experience with replacing these in power amplifiers?? My last output stage blew up after 5 years with no real cause, I assume some hf instability was the cause...
Answers are much appreciated, thanks
I found some Hexfets (irfp240 and Irfp9240) to replace the sj162/sk1058 (which obviously replace sj50/sk135). However I also read that they are not always suitable as replacement. They are a lot cheaper however (I can get them for around $2) and certainly worth a try. Does anybody have some experience with replacing these in power amplifiers?? My last output stage blew up after 5 years with no real cause, I assume some hf instability was the cause...
Answers are much appreciated, thanks
Exicons...
Hi all,
For any late viewers of this thread: Don't forget the Exicons. As an alternative to genuine Hitachi's people could also look here for replacements manufactured by Exicon/Profusion:
Profusion Exicon site
Click down to:
Equivalent chart
I have no direct experience with these, but the datasheets resemble the Hitachi sheets, and ratings seem to surpass the SJ50/SK135 ratings.
Kind regards, AudioGuy.
Hi all,
For any late viewers of this thread: Don't forget the Exicons. As an alternative to genuine Hitachi's people could also look here for replacements manufactured by Exicon/Profusion:
Profusion Exicon site
Click down to:
Equivalent chart
I have no direct experience with these, but the datasheets resemble the Hitachi sheets, and ratings seem to surpass the SJ50/SK135 ratings.
Kind regards, AudioGuy.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.