Yep. And it was American Airlines. And I drank so many (nearly frozen) Vodka shots along with the blinis, that I awoke in the middle of the night with an urge to barf. Fortunately I was able to sip enough water to dilute the stuff ;-)
Those 747 landings are pretty amazing.
...Jim Thompson
This was in '93-'94.
...Jim Thompson
-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |
Jim Thompson skrev:
lol, I can't imagine even First Class making 10 hours across the atlantic something you would do for fun ;)
-Lasse
The food was quite good, and most of it was "night".
My standard trick to Germany was to leave late afternoon from Phoenix, arrive Frankfurt at 8:30AM, then drive straight to Buhlertal and work all day.
After dinner I was definitely ready to sleep on a German schedule.
Returning, I did just the opposite, arrive Phoenix just in time for dinner, eat, drink, go to bed, back to Phoenix time in one cycle ;-)
...Jim Thompson
Jim Thompson skrev:
snip
amazing, the few time I've been flying with AA across the atlantic they served something I wouldn't give to a dog :)) Lufthansa's generally has pretty good food
I've been on that Frankfurt-Phoenix flight quite a few times, they used to have a 747 everyday , but now it is no more, so now it's usually via Denver:(
-Lasse
In '93-'94 when I was flying to Germany, it was Phoenix to Dallas, then Dallas to Frankfurt.
...Jim Thompson
Hello Robert,
As John said, Schottkys are very fast in recovery. Look at the ONSemi MUR/MURS series depending on whether you'd like SMT or through-hole.
Regards, Joerg
Hello Robert,
It should be feasible. If you use a fixed frequency oscillator as you already do that takes the first hurdle. Now you have to make sure your loop cannot fall into slight instabilities where it starts "pumping". For example, if your pulses cut in and out that will modulate the drive signal and become audible with the motor as the "sepaker". You could look at the (filtered) feedback signal with a scope to see how that behaves.
Then there are noises inherent to the motor construction that depend solely on speed. There is next to nothing you can do about those other than selecting a different motor.
Once I got a ride in a Ford Model A. Big surprise: This old car is surprisingly quiet on the outside but quite noisy inside. But it offers a lot more legroom than many modern cars.
Regards, Joerg
Hello Jim,
If that was on American Airlines, did you get the caviar on blini?
Once I snoozed so soundly in a first class seat that I didn't feel the landing in Europe and they had to wake me up. All the other passengers were already gone. Pretty embarrassing.
Regards, Joerg
Hello Jim,
Didn't know AA still uses those. My flights were usually on a 767.
Regards, Joerg
I read in sci.electronics.design that Joerg wrote (in ) about 'Whining PWM motor control', on Fri, 19 Aug 2005:
Because it's a tin box with no damping on the panels.
... which have 75 mm or so of insulator/absorber on every surface.
When we were students, my friend and I quietened our Austin 7 van quite a bit with internal insulation. Kept us a bit warmer, too.
That's not what's happening. The current limit is not resetting the FF during each 50us period, so the clock just re-clocks the '1' and the MOSFET stays on continuously until the current limit is reached. All you are doing is granulizing time into 50us intervals at which you enable the MOSFET to make an off-to-on transition. This is not PWM, it is some kind of quasi PFM half-hysteretic thing...
You could if set triggering to DC coupled, NORMAL mode, negative transition, on the current sense ramp channel, adjust timebase to something like 500us/DIV or more, and then decalibrate the timebase until you get a steady display.
Yep- sounds like that's what's happening.
Yes there is no sync between ON and OFF, so everything is random. The OFF time can be anywhere between 0 and 50uS, making the drop in motor current during flyback anywhere from 0 to 0.25A.... and that makes the ON time vary from 0 to 2mS (roughly).
On Sat, 20 Aug 2005 02:30:14 GMT, Fred Bloggs wrote in Msg.
yeah. It drives the woman who has to work around the device half-hystertic as well. I'll now go for a somewhat less direct approach; i.e., generate the PWM with a sawtooth oscillator and control duty cycle via low-passed current sense voltage. In other words, I'll throw a few compoents in the general direction of an SG3524 and see what happens.
Many thanks for all the suggestions everybody. I'll be back if something goes wrong.
robert
You could probably get a suitable diode out of the secondary rectifier of a PC power supply, if you have a few of those in your junk box.
To me it looks like your current comparator has some hysteresis, or perhaps the gate drive current of the MOSFET or the reverse recovery current pulse of the 1N4001 is affecting the current sense signal.
Is the FET turning on for a few nanoseconds every cycle, or is it really staying off for several cycles at a time? Probe the gate voltage with a fast scope.
Also make sure there is no deliberate positive feedback (hysteresis) in your comparator.
Chris
Have something to add? Share your thoughts — no account required.
Ask the community — no account required