Pages: 1 2 [3]   Go Down

Author Topic: 2N2222A TO-18 vs TO-92  (Read 878 times)

G4AON

  • Member
  • Posts: 2178
Re: 2N2222A TO-18 vs TO-92
« Reply #30 on: July 11, 2021, 01:19:00 AM »

The often used 2N3904, a favourite of Wes W7ZOI, is OK for DC and lower HF. For use above 10 MHz the MPSH10 is also a similar low cost but has double the fT (650 vs 300).

The GQRP Club has the MPSH10 available to members at 10 for 80 pence. Their 2N3904 is 50 pence for 10.

http://www.gqrp.com/sales.pdf

Regarding power dissipation, a cheap eBay “ear” thermometer can be useful. I use the type with a removable front cover that when in place switches the range to “wide”, up to well over 50C. Glue the switch to permanently switch it to wide range and use the narrow ear size opening close to components to measure their temperature. Note, metal can transistors need a black area from a marker pen on the side of the can in order to measure them with an infrared thermometer. Not a bad investment for 9 GBP delivered.

73 Dave

Logged

G0HZU

  • Member
  • Posts: 345
Re: 2N2222A TO-18 vs TO-92
« Reply #31 on: July 14, 2021, 03:46:41 PM »

Interesting stuff about the ear thermometer method. Is there a preferred model for this? I spotted a few on UK ebay for £10-£12.

The 2N3904 and MPSH10 are nice transistors but care with the circuit layout is definitely needed with the MPSH10. The quoted Ft may only be 650MHz at (say) 4mA and 10Vce on the datasheet but oscillation is still possible up around 1800MHz or so. VHF transistors like this are often at their most unstable when used as a common base RF amplifier. Even with a short and direct soldered base connection to a solid ground plane the MPSH10 or the SMD version the MMBTH10 can go unstable up at 1800MHz or so unless some form of stopper resistance is included or the rest of the amplifier layout is very tight and direct.

It may be the case that many people don't realise the instability is there because they won't have the test gear to display signals up at 1800MHz. However, some of the cheap SDR receivers can be used as a low cost spectrum analyser and these can work up to 2GHz. It probably is still possible for an MPSH10/MMBTH10 to self oscillate at about 2GHz but it is very unlikely to happen I think. Instability somewhere in the range of 1.2-1.8GHz is much more likely.
« Last Edit: July 14, 2021, 03:53:15 PM by G0HZU »
Logged
Regards
Jeremy

G4AON

  • Member
  • Posts: 2178
Re: 2N2222A TO-18 vs TO-92
« Reply #32 on: July 14, 2021, 04:03:55 PM »

The ear thermometer that I use is this type:
eBay item 294247559306
They are generic “4 in 1” thermometers. When the front cover is in place it switches the range to measure room and food temperatures. I glued the switch as I only use it to measure components.

73 Dave
Logged

G0HZU

  • Member
  • Posts: 345
Re: 2N2222A TO-18 vs TO-92
« Reply #33 on: July 14, 2021, 04:31:40 PM »

Thanks! I've got one of the popular Flir thermal cameras here but it is quite big so often I can't measure circuits that are tucked away inside an enclosure. Like you I have to do similar tricks with black paint to get reliable readings on some surfaces.
Logged
Regards
Jeremy

G0HZU

  • Member
  • Posts: 345
Re: 2N2222A TO-18 vs TO-92
« Reply #34 on: July 14, 2021, 05:38:47 PM »

I just had a play with a negative resistance oscillator circuit using a 2N3904 (genuine Fairchild part purchased from Farnell) and I managed to break the 1GHz barrier with it! This did require a fairly high Vce voltage (15V) but it didn't take much effort to make it oscillate at 1GHz and this was cross-checked on a spectrum analyser.

I'm not sure I'll manage to see 2GHz from a MPSH10 oscillator but I suspect I can get quite close.
Logged
Regards
Jeremy

KH6AQ

  • Member
  • Posts: 9292
Re: 2N2222A TO-18 vs TO-92
« Reply #35 on: September 08, 2021, 08:18:40 AM »

Speaking to the difference in thermal resistance, these two datasheets for the TO-18 and To-92 specify higher junction-to-ambient thermal resistance for the metal case. Note that these are two different manufacturers and the thermal test methods are not described.

2N2222, TO-18 metal case, junction-to-ambient thermal resistance = 300 deg C/W, junction-to-case = 83.3 deg C/W


PN2222, TO-92 plastic case, junction-to-ambient thermal resistance = 200 deg C/W, junction-to-case = 83.3 deg C/W


https://media.digikey.com/pdf/Data%20Sheets/ST%20Microelectronics%20PDFS/2N2222.pdf

https://media.digikey.com/pdf/Data%20Sheets/ON%20Semiconductor%20PDFs/PN2222.pdf
Logged

W9IQ

  • Member
  • Posts: 8866
Re: 2N2222A TO-18 vs TO-92
« Reply #36 on: September 08, 2021, 08:48:31 AM »

Speaking to the difference in thermal resistance, these two datasheets for the TO-18 and To-92 specify higher junction-to-ambient thermal resistance for the metal case. Note that these are two different manufacturers and the thermal test methods are not described.

2N2222, TO-18 metal case, junction-to-ambient thermal resistance = 300 deg C/W, junction-to-case = 83.3 deg C/W


PN2222, TO-92 plastic case, junction-to-ambient thermal resistance = 200 deg C/W, junction-to-case = 83.3 deg C/W


https://media.digikey.com/pdf/Data%20Sheets/ST%20Microelectronics%20PDFS/2N2222.pdf

https://media.digikey.com/pdf/Data%20Sheets/ON%20Semiconductor%20PDFs/PN2222.pdf

Here is a nice photo that shows why that is the "case":


Source: https://en.m.wikipedia.org/wiki/TO-18

- Glenn W9IQ
Logged
- Glenn W9IQ

God runs electromagnetics on Monday, Wednesday and Friday by the wave theory and the devil runs it on Tuesday, Thursday and Saturday by the Quantum theory.

KH6AQ

  • Member
  • Posts: 9292
Re: 2N2222A TO-18 vs TO-92
« Reply #37 on: September 08, 2021, 02:22:33 PM »

I can calculate the thermal resistance of the die-to-case from the picture but based on the spec and the surface areas I suspect the dominant difference is the case-to-air thermal resistance. The junction-to-case thermal resistance is specified the same (83 deg C/W) for both package types. The die to board (via the leads) thermal resistance may be the same too. That leaves the surface area of the packages as the variable. The plastic case has twice the surface area. 

The total package areas are:

TO-18, 0.45 sq inches
TO-92, 0.21 sq inches

Neglecting the lead-to-board thermal resistance we have the die-to-package and package-to-air. Using 40 deg C/W for the package-to-air we get.

TO-18, 83 + 40/0.45 = 172 deg C/W, spec is 200 deg C/W

TO-92, 83 + 40/0.21 = 273 deg C/W, spec is 300 deg C/W
« Last Edit: September 08, 2021, 02:35:02 PM by KH6AQ »
Logged

W9IQ

  • Member
  • Posts: 8866
Re: 2N2222A TO-18 vs TO-92
« Reply #38 on: September 08, 2021, 02:35:37 PM »

I can calculate the thermal resistance of the die-to-case from the picture...

How do you go about calculating θjc from the picture?

...but based on the spec and the surface areas I suspect the dominant difference is the case-to-air thermal resistance.

That is true since:

     θjajcca

Then realize that θca is not strictly a function of surface area. Take another look at the picture to understand the flaw in the remainder of your math.

- Glenn W9IQ
« Last Edit: September 08, 2021, 02:40:39 PM by W9IQ »
Logged
- Glenn W9IQ

God runs electromagnetics on Monday, Wednesday and Friday by the wave theory and the devil runs it on Tuesday, Thursday and Saturday by the Quantum theory.

KD7RDZI2

  • Member
  • Posts: 689
Re: 2N2222A TO-18 vs TO-92
« Reply #39 on: September 20, 2021, 01:00:32 PM »

To improve the power on higher bands from 20 to 10m you may change, if I remember correctly reducing the capacity of the cap from the emitter to ground of the driver. My ver.3 had less than 1w on 10m and it increased to over 2.5w. But be aware that if you want to use it on SSB to add a high pass filter for 10m to mitigate the issue of spurs. I would not used it on 21Mhz SSB at all. On CW as you probably know there is no such issue.

Hi,

Yes, with the 5109's, you just need one to replace two of the 2222a's.

The group is kind of a mess with many people trying things randomly and getting random results.  To me it's not really worth digging into all the threads.  Just go to the wiki page and down to the very last link: 

https://groups.io/g/BITX20/wiki/RF-power-and-gain

The changes here were done by Allison KB1GMX who is an expert EE and put a lot of effort into researching and testing the changes (I've already done the LPF relay re-arrangement mod she designed and it worked perfect at getting the harmonics down to legal levels).

There are basically two areas that need fixing:  Q90 biasing change to help tamp down the gain on the lower bands, then replacing all the 3904's which can't give the gain needed on higher bands.  The page also suggests changing Q90 to something with a better Ft because it is running kind of hot, and suggests good replacements.

73,  Mark
Logged
Pages: 1 2 [3]   Go Up