The Johnson counter is OK, but the divide-by-2 flop on the end could start out of phase if the clear goes away just as a clock hits.
Slow RC glitcher clears, and async clocks, are sloppy and dangerous design. One is obliged to prove that there are no hazards, instead of demanding that other people find them. It's easier to just design safe synchronous stuff.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
Out of phase with what? The Johnson counter is a divide by three and the bistable is a divide by two - they do change state on the same edges, but not always at the same time. This is just more pretentious bullshit.
But you missed the 11nF capacitors. Don't do much analog design these days?
You've definitely failed to cover yourself with glory here.
--
OK, then, with respect to what you seem to have in mind, let's
populate my circuit with real parts of your own choosing and work out
a solution for what you seem to think an edge rate violation might be.
Are you up for it?
--
True, but one may not always have that luxury available, plus there's
the synchronous reset to do, as well.
In deference to wise councel, I've changed my design from asynchronous
to synchronous reset and used an RC with a time constant several
clocks long on the "D" input of a dflop to sync both the leading and
the trailing edges of the reset with the clock.
Sheeeesh! Is Larkin saying he can't design a filter without a handbook or software ?:-)
I was designing (and successfully manufacturing) active filters BEFORE there were simulators. ...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 |
I love to cook with wine. Sometimes I even put it in the food.
I could build a doghouse without power tools. But why should I?
Sure. I was designing modems in 1969, with delay-equalized bandpass filters. But I mislaid my slide rule, so I have to use computers now.
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
I have that. Good book. The filter "designs" are mainly looking up prototypes in the tables and normalizing them, which is fine. I wrote a PowerBasic program to do the normalization math for LC filters, which helps when juggling to hit available values.
He has some good all-pass stuff.
But I still prefer to use a filter design program when I can, which is almost always nowadays. Saves a lot of typing.
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
Williams and Taylor put me onto the Sallen and Keys with some gain, which does make the capacitor values quite a bit more flexible.
It does cover more variations of filters than most. I found the "equi- ripple approximations to linear-phase filters" unexpectedly handy about twenty years ago - the classical version had too much gain at the front end, but the equi-ripple approximation came out a lot more practicable.
Without the divide-by-2, the RC glitch reset is safe. Only one flop gets set on the first clock, the first flop of the ring counter. And if it misses, it will fire on the next clock.
The filters still aren't flat, so if you change the clock frequency the amplitude changes a lot. With two opamps, you could do a 4-pole Butterworth or Chebychev filter and get nicer sine waves, flat over an octave maybe. That 3rd harmonic is the killer.
There's probably some resistive mixing that could be done from the available logic outputs that would reduce the 3rd harmonic.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
How about, because they are not necessary and can be expensive? Very especially if you do not know to either; how to value shop, or use possibly expensive tools. (Or possessively worse get crippled by using them: e. g. SPICE)
No, you can't have any of mine. I still know how to use them.
You think he designs around ideal opamps? I can't buy them in SOT23 packages.
I'm an engineer. I design a lot of electronics. Given the choice of doing things the hard way, or doing them the easy way, I cheat and pick the easy one. I'm not sure I can still do long division.
I used to design filters by hand. I used to lay out PCBs with tape and mylar. Neither was a lot of fun.
Latest laser driver board; did the layout on this one meself!
formatting link
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
Real Men love power tools. How do you drill holes? Use an old brace-and-bit? One of those hand-crank bevel gear drills?
Very
Spice saves a lot of time, sometimes. I just ran a bunch of sims on the LTM8023 switcher module, looking for sub-cycle oscillations. The LTC macromodel isn't super good, and the sim doesn't quite match reality, but it was educational. Looks like we need to stock up on polymer aluminum caps.
Well, I have a couple of HP35s that still work. And I do a lot of math in my head, which I still know how to use.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
'Real Men' are craftsmen with whatever tools they can find to use. Some of the finest things ever made were done with the simplest of tools.
As far as a drill? I have several drill presses, including a 'Cameron Precision'. At least a half dozen cordless drills, air drills, Dremmel, and their clones and I had a brace & bit that was very handy for drilling holes in power poles where there was no power to run a 1/2" power drill. You can't get through one hole with most cordless drills when drilling that treated wood. You also need special drills to cut the wood, and not burn up.
I have spent hours with files to make metal parts, when a milling machine wasn't available or to repair damaged power tools that could have been used to make that part. I just repaired the embedded controller for a pipe bender for a local factory. A replacement was over $700 and they would have been down over a month, waiting for the replacement from the OEM in Italy. they can't ship any product with that one machine down. The repair costs were under $7. There is no documentation available, other than the mechanical parts & a list of the modules. It uses a VFD a three phase motor and a Rabbit Semiconductor module, plus their custom display & keyboard. Their part is poorly designed, and was damaged because some idiot put a Ni-Cad battery on the interface board where it ate the traces off the board when it leaked. The first board worked for eight years. A brand new replacement only lasted 13 months. Like a lot of computer motherboards in the 286-486 days, it was designed to self distract.
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.