The Holy Grail Follower Output Stage

Jump to Latest
#521
I looked around a while back out of curiosity and although the FQA28N15 were in stock at Newark and Rochester, I didn't find any trusted sources that had FQA36P15.

Does anyone know if "Component Sockers USA" a trusted source? Octopart shows they have stock along with a bunch of less trustworthy options.
 
#522
I tested the fake MOSFETs with a basic component tester and the values obtained are similar to those of genuine MOSFETS (image below). But, from what I understand, fake components can perfectly mimic the 'behavior' of genuine components, the problem with these fake components lies in their reliability.
Furthermore, the visual inconsistency between the received MOSFETs stands out, and I don't believe that Fairchild had such poor quality control.
This is why I don't think it's worth investing time in taking more measurements when it seems obvious (to me) that these are fake components. Even if those measurements were within expected values, I would still think that these were counterfeit components.

PS: I'm only referring to my MOSFETs, the same doesn't apply to @myleftear's.

4.jpg

@Vunce
It seemed like a good deal. 😢
As a precaution, I just ordered a set of 10 MOSFETs because I was already anticipating the possibility of this outcome.
Fortunately, the loss was not significant. 😆
 
#524
FYI, I tested the gm of the 10 FQA28N15 Fets I received from Polida. They were all between 4.0 and 4.7. They did have that different look on the front of them compared to the FQA36P15. I will see how they perform in an amp one of these days.
 
#525
FYI, I tested the gm of the 10 FQA28N15 Fets I received from Polida. They were all between 4.0 and 4.7. They did have that different look on the front of them compared to the FQA36P15. I will see how they perform in an amp one of these days.
Click to expand...

The ones I purchased from Polida the FQA36P15 look like the picture. I'm thinking the FQA28N15s are not genuine. All the markings on this are exactly the same.

I may look at picking up some FQA28N15s from a known source.
 
#527
I finally obtained some Harris IRFP9140 FETs and tested them in the Holy Grail Follower.
The image shows distortion sweeps for the left and right channels using randomly selected pairings of IRFP140/9140 FETs.
Bottom line:
  • H2 is about 20dB above H3 and has positive phase relative to H1. H2 phase is easily inverted by using In- and reversing the speaker connections.
  • In A/B listening tests I cannot hear any difference from well selected FQA28N15/FQA36P15 FETs.
1714431378648.png
 
  • Like
Reactions: DIYHarry, Samuel Jayaraj, Botte and 3 others
#529
Analogico said:
A word of warning
I ordered a set of 10 MOSFETs (28N15) from this eBay seller. When I saw the image (see below) posted by @myleftear I was quite confident that I could receive potentially genuine MOSFETs, but unfortunately the articles I received seem obviously counterfeit.
Apparently, the seller may have genuine FETs as well as fake ones.
In other words, buying from this seller seems to be a game of chance.
Do business with this seller at your own risk!

@myleftear MOSFETs (which I assume @lhquam got from Polida's store)


MOSFETs I received from Polida (visual inconsistency between the various items is significant)


For reference genuine onsemi 28N15 MOSFET
Click to expand...
The FETs I received from Polida had the same "funky" weld on the legs as though they had been scavenged from something and made available for resale. Various manufacturer markings seem to suggest that they came from different fab lines and/or production batches.
I have so far found no problems with any of these FETs. The FQA28N15 FETs seem to be consistent with the same FETs that I purchased about 10 years ago. The FQA36P15 FETs have a wider variability in their gm, but otherwise have performed well.

Go ahead and do some simple tests on them. You would like to see gm of the Ntype between 4 and 4.5, the Ptype between 3.6 and 4.5 or so.
 
  • Like
Reactions: DIYHarry and ACnotDC
#530
I got two silkscreen versions from @lhquam (and thus from the same eBay seller) and measured both. Seem to be legit match regardless of silkscreen based on measurement. "New" is the top one with unbroken numbers. "Old" is the middle with the break for the number 8, Q, B, and A. I also added the Onsemi version.

20240430_183046.jpg

20240430_183159.jpg

20240430_190119.jpg


Here's a picture @Toys4Boys provided showing what he purchased from a reputable source for the OnSemi FQA28N15 (Newark) just so it's all in one place.
Toys4Boys Onsemi FQA28N15.jpg
 

Attachments

  • Id-Vds Compare New vs Old vs Onsemi FQA28N15.JPG
    Id-Vds Compare New vs Old vs Onsemi FQA28N15.JPG
    213.9 KB · Views: 95
  • Vgs Compare New vs Old vs Onsemi FQA28N15.JPG
    Vgs Compare New vs Old vs Onsemi FQA28N15.JPG
    206.7 KB · Views: 101
Last edited:
  • Like
Reactions: DIYHarry and Botte
#537
lhquam said:
I finally obtained some Harris IRFP9140 FETs and tested them in the Holy Grail Follower.
The image shows distortion sweeps for the left and right channels using randomly selected pairings of IRFP140/9140 FETs.
Bottom line:
  • H2 is about 20dB above H3 and has positive phase relative to H1. H2 phase is easily inverted by using In- and reversing the speaker connections.
  • In A/B listening tests I cannot hear any difference from well selected FQA28N15/FQA36P15 FETs.
So after all Harris or IRF 9140/240 combination can be used for the Holy Grail Follower, perhaps with lower bias currents. Is that correct?
Click to expand...
 
#538
So after all Harris or IRF 9140/240 combination can be used for the Holy Grail Follower, perhaps with lower bias currents. Is that correct?
Click to expand...
Post 557 shows distortion sweeps for the pairing of IRFP140 (?) with and IRFP9140 made by Harris. Those sweeps are with 12dB negative global feedback. The distortion levels are quite low into an 8 Ohm load, but not so good in to a 4 Ohm load.

I will discuss this further in a future post.
 
You must log in or register to reply here.
Top Bottom