454 Fireworks Followup


Sean Turner
 

Quick update: I was able to remove Q1197 from the +75V regulator circuit quite easily by loosening the power transformer and taking the rear panel off to go in through the fan opening. Threw that bad boy on the 575 and it is definitely shorted!

According to the manual, this is selected from 2N3442. Anyone have a spare one or two of those laying around they would be willing to part with, that are known good, before I take my chances on ebay? Willing to pay a reasonable fee + your cost to mail.

Thanks!

Sean


 

Sean,

I've been trying to find what the selection criteria were for Q1197, which is Tek part number 151-0209-00, if I am not mistaken. The service manual does say that it is selected from 2N3442, but when I consult the Common Design Parts Catalog I do not find 151-0209-00 listed under selected transistors. Further, in the section that discusses (and tries to dissuade engineers from using) selected parts it sounds like a selected part should have a part number starting with 153- rather than 151-. It also sounds like there are a population of 151-****-89 parts that are still considered "selected" in the sense that while they have failed the selection criteria, they have been checked to be free of other faults, and so are considered higher quality parts.

Could it be that Q1197 is from the latter population of selected parts and that the -89 suffix was not yet in use when the 454 was designed? Maybe any old 2N3442 would suffice, so long as it was working?

-- Jeff Dutky


Sean Turner
 

Jeff,

Yes, indeed that is the part number on the transistor I removed.

I'm figuring a plain old 2N3442 that is free of defects (I have curve tracers and the test fixture for this case style) should work fine. I want to give it a try at any rate. Apparently this part is still manufactured by Microchip/Microsemi, but it isn't stocked at the usual suspects and is only available in bulk quantities for absurd prices and long lead time. So my options are someone here or ebay.

Sean

On Wed, Mar 24, 2021 at 09:01 PM, Jeff Dutky wrote:


Sean,

I've been trying to find what the selection criteria were for Q1197, which is
Tek part number 151-0209-00, if I am not mistaken. The service manual does say
that it is selected from 2N3442, but when I consult the Common Design Parts
Catalog I do not find 151-0209-00 listed under selected transistors. Further,
in the section that discusses (and tries to dissuade engineers from using)
selected parts it sounds like a selected part should have a part number
starting with 153- rather than 151-. It also sounds like there are a
population of 151-****-89 parts that are still considered "selected" in the
sense that while they have failed the selection criteria, they have been
checked to be free of other faults, and so are considered higher quality
parts.

Could it be that Q1197 is from the latter population of selected parts and
that the -89 suffix was not yet in use when the 454 was designed? Maybe any
old 2N3442 would suffice, so long as it was working?

-- Jeff Dutky


 

Sean,

Or Amazon: https://www.amazon.com/s?k=2N3442

I'm not sure Amazon is really any more reliable than eBay, but just so you know your options.

I did find 114 count 2N3442 and 951 count 2N3442G in stock at Mouser, both available in qty 1 for about $6.75 each, so you have an option of a reputable source, but it comes with a price (the stuff on Amazon is about $2 each, but you get what you pay for, maybe)

-- Jeff Dutky


Sean Turner
 

Jeff,

Thanks, maybe I'll go with Mouser. I order stuff from them with some regularity, and that's not a horrible price IMO. Amazon is right out...too much counterfeit garbage. Though it might be fun to get some of both and compare their characteristics.

Sean

On Wed, Mar 24, 2021 at 09:56 PM, Jeff Dutky wrote:


Sean,

Or Amazon: https://www.amazon.com/s?k=2N3442

I'm not sure Amazon is really any more reliable than eBay, but just so you
know your options.

I did find 114 count 2N3442 and 951 count 2N3442G in stock at Mouser, both
available in qty 1 for about $6.75 each, so you have an option of a reputable
source, but it comes with a price (the stuff on Amazon is about $2 each, but
you get what you pay for, maybe)

-- Jeff Dutky


Richard R. Pope
 

Sean,
I would use Mouser or Digikey. I trust both of them. I don't trust E-bay or Amazon for electronic components.
GOD Bless and Thanks,
rich!

On 3/25/2021 12:00 AM, Sean Turner wrote:
Jeff,

Thanks, maybe I'll go with Mouser. I order stuff from them with some regularity, and that's not a horrible price IMO. Amazon is right out...too much counterfeit garbage. Though it might be fun to get some of both and compare their characteristics.

Sean

On Wed, Mar 24, 2021 at 09:56 PM, Jeff Dutky wrote:

Sean,

Or Amazon: https://www.amazon.com/s?k=2N3442

I'm not sure Amazon is really any more reliable than eBay, but just so you
know your options.

I did find 114 count 2N3442 and 951 count 2N3442G in stock at Mouser, both
available in qty 1 for about $6.75 each, so you have an option of a reputable
source, but it comes with a price (the stuff on Amazon is about $2 each, but
you get what you pay for, maybe)

-- Jeff Dutky



Michael W. Lynch
 

On Wed, Mar 24, 2021 at 11:01 PM, Jeff Dutky wrote:


I've been trying to find what the selection criteria were for Q1197, which is
Tek part number 151-0209-00, if I am not mistaken. The service manual does say
that it is selected from 2N3442, but when I consult the Common Design Parts
Catalog I do not find 151-0209-00 listed under selected transistors. Further,
in the section that discusses (and tries to dissuade engineers from using)
selected parts it sounds like a selected part should have a part number
starting with 153- rather than 151-. It also sounds like there are a
population of 151-****-89 parts that are still considered "selected" in the
sense that while they have failed the selection criteria, they have been
checked to be free of other faults, and so are considered higher quality
parts.

Could it be that Q1197 is from the latter population of selected parts and
that the -89 suffix was not yet in use when the 454 was designed? Maybe any
old 2N3442 would suffice, so long as it was working?

-- Jeff Dutky
I vaguely seem to recall a discussion from some time ago regarding a similar type of power transistor, in another instrument.. Something about epitaxial (high gain) vs hometaxial (low gain) construction with a lower gain hometaxial construction being the "Desirable" characteristic in the particular application? I just do not remember all the details. Or perhaps I am just wrong?

I see that the TEK Common Parts list calls for an HFE of 20 and the 2N3442 shows an HFE range of 12 to 72. Perhaps this memory above applies and TEK selected those parts with a fairly low HFE?

Just a thought.
--
Michael Lynch
Dardanelle, AR


toby@...
 

On 2021-03-25 10:48 a.m., Michael W. Lynch via groups.io wrote:
On Wed, Mar 24, 2021 at 11:01 PM, Jeff Dutky wrote:


I've been trying to find what the selection criteria were for Q1197, which is
Tek part number 151-0209-00, if I am not mistaken. ...

-- Jeff Dutky
I vaguely seem to recall a discussion from some time ago regarding a similar type of power transistor, in another instrument.. Something about epitaxial (high gain) vs hometaxial (low gain) construction with a lower gain hometaxial construction being the "Desirable" characteristic in the particular application? I just do not remember all the details. Or perhaps I am just wrong?
That was probably about the 2N3055(A), it comes up regularly.


--Toby


I see that the TEK Common Parts list calls for an HFE of 20 and the 2N3442 shows an HFE range of 12 to 72. Perhaps this memory above applies and TEK selected those parts with a fairly low HFE?

Just a thought.


Michael W. Lynch
 

On Thu, Mar 25, 2021 at 10:11 AM, <toby@telegraphics.com.au> wrote:


That was probably about the 2N3055(A), it comes up regularly.

--Toby
OK, then I was not imagining that this discussion had occurred. I was wondering if this could be a similar situation? Your thoughts?

Thanks!

--
Michael Lynch
Dardanelle, AR


John
 

I was going to offer to send you the "same device" from a 453, only to discover that it's a 2n3441, i.e. smaller TO 66 case! (The 453 has a shunt resistor across the transistor which is either to assist self-starting or reduce device dissipation, or both).
Unusually, I can buy the 2n3442 much cheaper here in the UK, but I guess postage out to the US would bring costs back up again.
https://cpc.farnell.com/unbranded/2n3442/transistor-npn-to-3/dp/SC06770
NB your post regarding the 575 testing prompted me to put the 2N3441's test to have a look.
I have two with Tek part numbers, and two I bought about a decade ago from I know not where (now) .
With no base current they al meet the 140V spec, but the non-Tek samples have a lower "knee" (c. 90V) where collector curves head north rapidly when passing collector current. I'll try to post some photos. Not sure what the mechanism is called, but I'd say the transistors couldn't be of much use in this region, at least as linear devices.
John