I've done that in Eudora. I'll have to figure out how to do it in Agent (v6). ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
Do you mean spell-checking the custom dictionary? That's needed occasionally, but spell checkers contribute to people using correct words in the wrong context, bit harder to fix.
Inverting the bridge+bulk cap and the pass transistor has nothing to do with start-up control.
It's just an elegant/simple way to not rise dissipation and drop line reg to 48dB or less, or loose 20dB of loop gain and 200uV/K of tempco, all of which is useful given the poor PSRR of the (simple design) active probes.
"fro" is in the "standard" dictionary which, in Agent, seems not to be editable :-( I have a support question in to Agent to see for sure. ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
I'm still waiting for elaboration on your beta claim. ...Jim Thompson
-- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at
formatting link
| 1962 |
Remember: Once you go over the hill, you pick up speed
Sounds like double-talk to me. Walk me thru the schematic showing this "fault". ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
If you depend on base current as the metric in a limiting protection circuit, you're introducing beta variation into the equation. This same beta variation has, in the basic design, already required that 'more than enough' base current is available for normal function - so the potential spread at limiting is doubled.....
Regulating or reducing current ripple into this node is a good way to improve PSRR, but this can be achieved without subjecting the entire topology to a brain fart.
I've done this kind of butterfly thing in small one-off projects, (all the way through....) where there was no limit to the idiocy I felt like including..... but I wouldn't suggest it seriously for a commercial design, using other people's time and resources and the 'blindly trusting' customer's money.
Secondary break-down current as illustrated in the SOA charts. These have a dual slope derating as the Vce increases.
At 24V and 25degC case temperature, the parts used for the main pass transistors in the 1122 probe power supply BOM are only safe to an amp (ie less than the 30W specified for the part in a strictly thermal derating).
In order to avoid it, the constant current limit of R4/Q3 and R15/Q8 has to be reduced somehow, as greater fractions of the input voltage show up across them.
The only method provided to do this is base current starvation of the darlington pass transistors, but only when the outputs are shorted or overloaded to each other. There is no reduction in base current supply if outputs are individually overloaded to ground.
When the protection is provided, it only occurs as the outputs approach voltages around 3xVbe. Even when completely shorted, how much current in Q1 do you think is still possible, thanks to the ~10mA of 'start-up current' flowing into the base of the PNP darlington pair?
Sheeesh! What a day! Several posters here trying to apply above Larkin-level BS.
The pass transistor passes SOA... WELL-INSIDE the DC SOA boundary. It's even still inside the DC SOA with an overload.
You should avoid "analyzing" from the hip. Try the pencil, paper and Ohm's law approach :-)
Maybe everyone has been confused by the unorthodox schematic representation? That happened during my Masters Oral... a professor couldn't recognize an upside down diff-pair. After that dumb-ass question, met by tittering from the rest of the committee, there was dead silence. Finally my advisor said, "OK, we're done. Let's go play golf." :-) ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
Addendum... at a dead short it's slightly outside the DC SOA, drawing
2A, so the 1A fuse should blow... it's labeled "normal". ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
I'm pleased with the way my KDE Knode spell-checker works - it doesn't "autocorrect" anything - when it sees a questionable word, it highlights it in this window, for example "Knode" above.
You're talking about 2N4818 and MJE521? With a 24V supply? A static limit around x amps?(.6/.35)
The SOA curve for DC begins to break into the secondary Isb slope at about 1.5A and 20V, again, at a case temperature of 25C.
With the imbalance in short circuit limiting behavior noted previously, the actual result of the rails shorting to each other is a negative supply voltage appearing on both rails (if nothing in the load clamps it).
"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...
I can verify that, indeed, with a dead short the 1A fuse does blow... I watched mine do so in perhaps 10 seconds or a bit less, which I believe is well within the spec for 2x overcurrent on a normal-blow fuse.
Legg shot from the hip. Me? I thought I was correct from just casual observation but, when Legg became obtusive ("Isb" my ass... even HSpice has no such creature), I simulated it (I had models for every device in the BOM).
In PSpice it was a piece-a-cake to also display SOA.
(I even have SOA macros for hot electron effects in CMOS short-channel devices :-)
Besides... How many years did hp make that particular supply without problems? ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...
That was probably PSpice rather than LTSpice, right? -- This has been a popular enough thread that I'm thinking it's worth the effort to enter it into LTSpice, say, tonight. I wasn't going to worry about the proper transistor models, since most people are probably mainly interested in how it works in general and are OK with believing HP designed within the SOA. :-)
Oh, plus it's a chance to draw it "upside-right!"
At least a decade, I think!
As far as I can tell, the main use of this power supply today is to power up Agilent 85024A FET probes (which have also been around for decades now) -- Agilent keeps the same power connector on, e.g., their current VNAs and spectrum analyzers, but if you want to use that probe with a Tek or Rohde & Schwarz, etc. unit, the 1122A here is still the way to go. (HP never seems to have any firm specs on how much power a probe is "allowed" to draw, but it does seem to be in the ballpark of at least 250mA per rail... which unfortunately gets you upwards of 7W, well in excess of what USB ports can provide... otherwise it'd be a cool little project to a build a "USB to HP FET probe power supply" widget...)
Yes, a tranny is a tranny... except for the high current parts. I have all the models if you want them. (I used 2N2222 and 2N2907 for the "generic" devices.)
...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
I'm sorry - I've lost track of this thread. Could someone clue a student in? Are we still discussing why the PS regulates the low side? Has the function of R7 and R8 been agreed upon? FOR THE LOVE OF GOD WILL SOMEONE FIND OUT THE ANSWER?! :(
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
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.