Maximum switching frequency for MJE13005 ?

Could some electronics guru please help ? I am trying to find out the maximum switching frequency fro the MJE13005 ? Both the Central Semiconductor and OnSemi datasheets do not provide any clear information -- "high frequency switching applications" -- but how high ? I know that the maximum switching frequency forthe 2N3055 is 4.2 MHZ. Any hints/pointers/ suggestions would be helpful. Thanks in advance.

Reply to
dakupoto
Loading thread data ...

daku

** No you don't.

** Try asking a question that has an actual meaning some day.

Amaze everyone.

.... Phil

Reply to
Phil Allison

There's no hard value. But why use that old, slow, low gain klunker? Why not a mosfet?

I

Where did that number come from? A 3055 is pretty slow, and can't do anything useful at 4 MHz. Ft is only 2.5.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Did you see the switching time specs? t_d t_r t_stg t_f

Note the collector (load) current and base current (I_B1 = on, I_B2 = discharge/turnoff) figures. The larger the ratio I_B2 / Ic, the faster t_stg is.

2N3055 might have a smigeon of power gain at 4.2MHz, given appropriate source and load impedances. There's no way in hell you've seen an authentic one* _switching_ that fast...

*I qualify this statement, on the off chance that some idiotic manufacturer put a screaming fast audio transistor inside one, e.g.

2SC3519
formatting link
which has an fT up to 55MHz at the right current!

In general, BJTs aren't useful for switching above maybe fT / 10. I've used MJE1300x series transistors at 200kHz before; switching losses will be too high for them to be useful much above that.

There is no hard frequency cutoff, only a derating due to RBSOA and switching losses.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

Small signal current gain cutoff frequency fhfe is 10kHz. (JEDEC registered number).

Plausible fear, but that would be a cruel trick- I bet a lot of old circuits would do 'undesirable' things.

--sp

--
Best regards,  
Spehro Pefhany 
 Click to see the full signature
Reply to
Spehro Pefhany

*MINIMUM* registered number! I suspect transistors that bad, haven't been made in a very long time, and can't even be made anymore.

This is why I explicitly avoid anachronisms like 2N3055 and 2N/PN2222 -- the specs can exceed the JEDEC listings by an unknown amount.

I've tested "2N3055"s in a circuit like this,

formatting link
at lower voltage (12V supply) and higher hFE (3 to 30 turns on the feedback/drive transformer). A 2N3055 from the '70s went slow as molasses (like you'd expect), one from the '90s worked just as well as a (presumed good) MJ15010, with something like 400ns switching edges.

They (2N3055s) might very well be a sale path for otherwise-good dice that fail to meet spec on a more expensive transistor type. Or, they make more on quantity by using the same (in-spec) dice in multiple parts.

The "good" ones also have low Vce(sat), like any other regular transistor.

(As for "bad" ones, I suspect you're more likely to get a 2N2222 die in your 2N3055, say if you order direct from China. Counterfeits are a real thing, but they're usually about shoddy packaging or inferior die size. But that's a whole 'nother can of worms.)

No thread is complete without pics -- take a look at this comparison of GP transistors:

formatting link
formatting link
formatting link

On Semi tends to care more about their transistors, though I forget if they spec their '2222's as if they're a fully in-house part, or if they follow the JEDEC spec and that's that.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

Yes, MOSFETs are definitely an option, but I "klunkers" like 2N3055 have done performed very well for our designs in the recent past.

Reply to
dakupoto

While it is very true that a Chinese fake devices barely meet the pre-defined device specifications, a number of local manufacturers in several countries provide genuine devices which, as expected, are more expensive.In India for example, two government owned electronics companies nabufacture and sell devices as 2N3055, MJE13005 etc., In these places, if you buy devices in bulk from wholesalers, they will specifically ask if you want the el cheapo ones or higher quality local makers.

Reply to
dakupoto

I am sorry for the typo, the maximum switching frequency for the 2N3055 is indeed 2.4 MHz.

Reply to
dakupoto

** But only on planet Daku.

.... Phil

Reply to
Phil Allison

My first buck switcher used a 2N3055. At 20 KHz. Must have been around

1970. Military, but similar to this one:

formatting link

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

That's Ft, the frequency where beta=1. It's not useful as a switcher at that frequency. 20 KHz or so is more reasonable.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

I consider parts like that good for circuits where the requirement is "this thing has some decent gain at the frequency of interest" and I don't really need to think about switching speed or saturation voltage.

That actually covers a lot of ground!

--
----Android NewsGroup Reader---- 
http://usenet.sinaapp.com/
 Click to see the full signature
Reply to
bitrex

If you cascode it with a low-voltage fast transistor, it could do some useful amplification even with beta below 1. There's too much dependence on external circuitry to set a hard limit, and there's no hint of operating point or 'switch' thresholds (can one apply a Baker clamp?).

As far as a 2N3055 is concerned, we're safe: that's a metal-case transistor that will never again be used in a new power-switching design. We hope.

Reply to
whit3rd

3055's aren't cheap any more. And not surface mount!

The first switcher I did, the switch stage was a small PNP, 2N2905 or something, driving a 2N3055 as a compound PNP. The 3055 never saturated, which made it relatively fast. The input was 24 volts, so another 0.6 drop didn't hurt much.

It was a pretty weird circuit, done long ago. Maybe I'll try to remember it and post it.

I used to be able to hear above 20KHz, so I ran it at 22K, just past my limit.

We had a visit yesterday from EPC, who make amazing chip-scale (unpackaged) high current GaN fets. Their customers are designing 50 MHz switchers, 2500x faster than my first switcher.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

I'm feeding that little audio transformer used as a step up converter 20kHz. I was hoping to hear a cool whine or something, but all I hear is silence.

I feel old..

----Android NewsGroup Reader----

formatting link

Reply to
bitrex

The JEDEC registration was for a hometaxial 'mesa' device, which was not built for speed, just ruggedness. They required a large periphery and added passivation (silicone etc) to avoid leakage on chemically etched edge detail of a larger die.

2N3055s have been manufactured in epitaxial 'planar' construction, at reduced cost, by most manufacturers, since the mid 80s. Higher frequency operation and degraded Isb (SOA and surge) ratings result in the smaller die.

For standard off-line switching, MJE13005 sees a practical base-driven brick wall above about 65KHz. Emitter-driven, as in a fet-cascode, this can be extended above 100KHz. Not all parts with this marking are created equal. You'd be better off buying a part that still has dedicated mfr support for bipolar power, if you must use bipolar switches.

RL

Reply to
legg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.