| Pages:
1
..
5
6
7
8 |
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Hey Tim , If that sim software is so good ,
then why didn't it flag that particular transistor ?
I caught it because I was curious about the
specs on the P channel device and looked it up .
Anyway it does simplfiy things considerably if
you can find complementary devices . I am scrounging
parts to build , and keeping it cheap
Seeing there are a lot of surplus high voltage N channel Mosfets available for cheap , I focused on them .
There are also some high voltage NPN transistors
available for cheap , but the beta really stinks for
the high voltage bipolars , so N channel Mosfets rule
in the performance versus cost scenario .
|
|
|
Twospoons
International Hazard
Posts: 1327
Registered: 26-7-2004
Location: Middle Earth
Member Is Offline
Mood: A trace of hope...
|
|
It depends on whether the breakdown characteristics are in the model for the fet. A sim is only as good as its models. The point is, it gets you in
the ballpark really fast. I just picked those fets at random  , because I
couldn't be arsed looking up the specs.
Helicopter: "helico" -> spiral, "pter" -> with wings
|
|
|
Lambda
National Hazard
Posts: 566
Registered: 15-4-2005
Location: Netherlands
Member Is Offline
Mood: Euforic Online
|
|
A Dirt-Cheap, High Current Power Supply
Here's yet another use for that pensioned-off AT computer that's gathering dust in the corner.
Got an old PC gathering dust somewhere? It mightn't be much good these days but it's power supply could be... especially if you want a high current
bench supply! This article tells you how to modify one - at very little cost!
Website:
http://www.siliconchip.com.au
A Dirt-Cheap, High Current Power Supply:
http://www.siliconchip.com.au/cms/A_30705/article.html
And for the printed version:
http://www.siliconchip.com.au/cms/A_30705/printArticle.html
Remark:
Apparently you can only access a limited amount of articles in full. Then you are requested to pay for the full article.
This is just to good to miss out on, so here you have the Rapidshare link to the full article:
Power Supply.rar (548 KB)
http://rapidshare.de/files/13312579/Power_Supply.rar.html
[Edited on 15-2-2006 by Lambda]
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Further revision of my earlier circuit idea has brought
this design process to a point where I think the device
is ready to build , after I do a final check of my calculations
to try to find any errors or see if any values may be adjusted
to any advantage in refining what I have now .
I have added the circuit which provides a full voltage
startup pulse , adjustable upwards to a defined
burst quantity of full voltage cycles , so that when the stirrer
is switched on at a minimal speed setting , the starting
torque is greatly increased to give a clean breakaway
of the rotor into rotation , in an adjustable way that
should prevent any overshooting of the set speed .
This power control circuit combined with the PSC motor setup
having a squirrel cage fan and eddy current loading on
the lower endshaft , as I have described earlier in the thread
will result in what I believe will be a magnetic stirrer
drive scheme having superior performance and durability ,
well beyond the performance of the top of the line products
which are supplied by the major manufacturers . And if this
belief proves to be true by the performance of the prototype
which I am building , then this is state of the art in the making
( and that wouldn't be a first at this forum either )
The original attachment here has been replaced with
a corrected schematic which deletes a capacitor filter across
the power switch , for reason that it could cause a problem .
See the posts below for more about this .
[Edited on 16-2-2006 by Rosco Bodine]
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
I'm just going to sit down, watch this thread and laugh when it lets out smoke.
Tim
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
There you go with the negative vibes ,
trying to jinx my brainchild with self-doubt
I'm scrounging parts together and I'll give an honest
accounting of how it works or doesn't work .....
the unvarnished truth , such as the case may be .
Honest , if it burns it will be my first smoker and
I've put together a few things over the years .
You younger folks were making fun of my arcwire crossings
on hand drawn schematics that went out of fashion
in the 1950's . I can't read the copyright dates in
the books I learned this stuff from because I forgot how
to read Roman numerals ! So you are probably right
in hinting I needed to catch up to the times . It took
me a good two hours to learn how to use the ECAD
software to draw " respectable " schematics .
So there now .....I'm a modern , even though I still
miss riding the DC3's and DC4's and Constellations
when they were in regular airline service ! I am a
pre " jet age " sliderule operator dragged kicking
and screaming into the age of the inkjet published
virtual drawing with a schematic tucked in my pocket
that I see no reason shouldn't work like I think it
will .
So if you see a glaring fault , please elaborate .
Do you think the trickle current through the 26.5K reactance
of the .01uF cap across the open switch will trickle charge the
RC timing circuit for the startup pulse , defeating it ?
If so the cap can be omitted , it was sort of a last minute idea
I really hadn't thought about very much .
I'll probably leave it off just to eliminate that potential gremlin . Probably it is better to put a varister across the motor to clamp any spike ,
and just leave the open switch open . Yeah , I think the varister makes more sense .....
so I'll change that detail .
Thanks for the suggestion  Why didn't I think of that
Earlier in the thread TransZorb was mentioned by densest .
Across the motor leads is probably the way to go here if
there is an appropriate voltage rating TransZorb for the task , and I'll defintely lose the cap on the switch because its a potential problem .
Update:
The attached schematic in my last post above has been updated with a corrected schematic deleting the anti-arcing capacitor which I had thoughtlessly
placed across the switch without thinking it through . A suitable value TranZorb is
recommended as an option for putting across the motor ,
but it may not be necessary . The power stage for this
device is hopefully rugged enough not to be vulnerable to
the transients associated with a small motor load . I tried
to make the vulnerable part of the Mosfets , the gates ,
as bulletproof as I could , knowing that the toggle switch
and full power setting or hot startup pulsing feature ,
would be used at times to jog the drive in on-off fashion ,
occasionally being used this way to shake a stirbar free which may be caught in a dense mass of crystals .
There are a few final changes which I am still making
in the circuit before I consider this a completed design .
So in no way consider the attached schematic above as being
a finished work . There are some revisions which
I will be posting with an updated schematic when the calculations are completed and the edits made for
the values of components I will use .
[Edited on 17-2-2006 by Rosco Bodine]
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
The attached schematic now shows the above mentioned
component revisions / additions . It is now closer to completion . In terms of arrangement of components
I believe the circuit is complete . However the values
for the components are not fully analyzed and checked .
So there may be some changes yet made , in the way of
further tuning and calibrating the system . I am mainly looking at optimizing the RC time constants for the startup pulse controller stage since that
is the last section added .
For you fellows who were following the circuit evolution
take a look at the timing circuit which biases Q4 .
The idea here is that on initial powerup Q4 is off and
allows the entire divider string of the control pots to
float to the potential of the positive rail , applying full
voltage reference to the gate of Q3 and followed from there
to the Output Mosfets Q1 and Q2 . While this is occuring
the capacitor in the zener regulated 10V charging path
is rising from 0 Volts towards the 1 Volt or so needed
to cause Q4 to switch , which takes the end of the divider
string for the control pots to ground , and makes their
settings effective on specifying V ref . to Q3 . The time
required for the capacitor to charge to the switching voltage
of Q4 will determine how many cycles of full power will
be delivered to the load . The larger 2.2 uF capacitor which
charges much more quickly to the full voltage of the bridge ,
acts as a virtual battery , which supplies power continuously to the 10V regulated timing circuit between cycles , so that
once Q4 has switched on , it stays switched on through the zero cross and into the next cycle , and the timer controlling
Q4 does not reset , until power to the bridge itself is interrupted and both caps bleed down .
[Edited on 17-2-2006 by Rosco Bodine]
|
|
|
densest
Hazard to Others
Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline
Mood: slowly warming to strain point
|
|
www.bestlabdeals.com/DATAPLATE_DIGITAL_PROGRAMMA_p/bmmp174.h...
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
@densest
That's a bum link . Anyway it doesn't matter what sort
of digital crutch you strap onto hardware to make up
for poor mechanicals . And on the well designed
equipment having well matched components dedicated
to the intended use , a purely analog and open loop
system is entirely sufficient . BTW I have something
like ten thousand dollars worth of commercially made
stirrers including top of the line Thermolyne and Electromantle , Fisher , and Corning , including some
brand new this years model stuff and a few pieces
that go back to 35 years ago . And for lab grade
instruments ......NONE of it is what it ought to be
in the magnets , the motors , or the speed controls ,
including the digital model . I know from studying the
equipment by tearing it down to its nuts and bolts and
cleaning and reassembling it , that I can design and build something way better quality and I mean to prove that .
I am still working on finalizing the design I have been
developing , but I don't know there isn't an easier way .
There could be . Did you look at that zener in the schematic
you posted , the one that was upside down ? Because I didn't really edit the circuit and study it through afterwards . But it seemed like a variation
on my original idea to use a bridge rectifier DC load to series regulate AC , and you had found a way to make that idea work , after I dumped that
idea and moved in the direction which brought me here to
the circuit I have now . So maybe I'm using two power
Mosfets when only one would do . I'll go back and take another look at it when I get this current design squared away since I have high confidence it
will work well . It's
always good to have more than one way of doing the same thing .
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
| Quote: | Originally posted by 12AX7
I'm just going to sit down, watch this thread and laugh when it lets out smoke.
Tim |
Hey Tim ,
Screw the SPICE simulation for this simple circuit ! .....
The conceptual and mathematical analysis and a discerning schematic reading is sufficient to say " this will work , so build the damn thing "
........ and that's from the old school , " Old Spice " sliderule jockey , SPICE illiterate
as I am . If I'm wrong , then I'll have to eat my share of humble pie when you fellows nail my ass to the wall
here with well deserved ridicule ......later ,
* if * the prototype burns . ( But it won't ) I'm not
worried about a design after the point where I can visualize
the intended result as a mathematical certainty ,
the predictions are over and it's time for proof .
I'm a great believer in perf board , legwire bending ,
and soldered connection " simulations " which run the way
you " figured " when you plug the damn thing in and throw the switch !
I look at this dedicated circuit which I have designed for a niche application , and without any personal bias from ego
afflicting me with any modesty either .......
my most objective opinion would be that obviously it was a genius who designed this little work of art
And the design is damn good too , absolutely beautiful .
If she was a girl , she would be one tough bitch ,
a little top heavy maybe ....but hey I like a girl who
sports a substantial package and puts out a little heat
So for all you jokers heckling and naysaying ,
I keep looking at this thing and looking for the fault ,
trying to spot the defect and I honestly don't see a damn thing wrong with it . In my opinion it will work and it will
work as intended .....so if you see something there in
that last schematic , which just doesn't add up or looks
wrong .....hey speak up and point it out to me ,
because the simulation I have been running on this thing
in my head keeps saying it absolutely will work in the
way it was designed and intended .
I still say this thing is a work of art , a beastie from
Planet Analog as it may be .....analog is not bad ,
when analog suits the application perfectly well ,
why complicate things going digital , and lose the
stepless smoothness you have with " hydraulic logic " ?
Have you ever flown a turboprop with a constant rpm
variable pitch propellor ? Smooth as silk , and no
digital logic involved . Similar regulation concept there
as what I am about here ......there is feedback and
governing , regulation inherent in the system as
" hardware logic " so the power feed just has to
feed power , and the mechanicals automatically
do what they are supposed to do with that power
they are getting .
And for those diehards who would insist on
adding digital rpm sensing and closed loop " servo locked "
operation .....well think about the advantage
there is about a system which already has a
great deal of inherent regulation , and how much
smoother and responsive it will be to amplified feedback
since the error correction required will be so much smaller
than for a system which fully relies on active external
regulation to maintain a set speed .
It's like the situation of a cruise control on an automobile
with a 2 liter engine approaching a grade , versus the
cruise control on an automobile with a 5 liter engine
approaching the same grade , the cruise control on
the car with the larger engine is going to have less
math to do and less error to correct to maintain the
speed of the system under the changing load .
No matter how good the cruise control , when
you are a passenger riding through hilly terrain ,
the vehicle with the larger engine is going to by
its nature provide the smoother ride .
That certainty is at the heart of my purpose of
using a specially configured drive motor in a
10 inch square stirplate which is a several times
" larger " motor in terms of its actual performance
than those deficient motors typically selected by
the commercial manufacturers , while it actually
uses no more electrical power than the junk they
are using instead .
What I am doing here is like the difference between
a " mini " and a Cadillac autobahn edition in terms of
inherent " reserve power " , available torque , ect .
I'm way too old to " go back to school " if I didn't learn
what I needed to know , to know this circuit will work
when I was in school thirty plus years ago .....
So I'm just going to build this thing and throw the switch
and see what happens
I should be taking bets on this thing so I could pick up
the extra money ......because if I had to charge for this
circuit design instead of making it a freebie here for
all my nerdie online counterpart friends .....it would be costing somebody about 200K for the R&D work carried
through to successful prototype for proof of concept .
I need a gold laminated pocket protector for this one ,
and a ball cap with gold embroidery which reads ,
" Captain Nerd " , and a silkscreened tie-dyed polo
shirt , with a pocket of course , carrying the logo
" Nerds Rule " . Then I would be a completed work
with my thick glasses However , I do have tortoise shell
eyeglasses frames instead of the ebony black , since I am a
" sport model " nerd
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
| Quote: | Originally posted by Rosco Bodine
| Quote: | Originally posted by 12AX7
I'm just going to sit down, watch this thread and laugh when it lets out smoke.
|
Screw the SPICE simulation for this simple circuit ! ...
|
What SPICE?
| Quote: | ...
The conceptual and mathematical analysis and a discerning schematic reading is sufficient . . . |
As a matter of fact I have been doing exactly this. I have never bothered to simulate any circuit. The last time I tried I found the software almost
impossible to use. I know there are freeware versions better than the one I had, but nonetheless, I could care less when I have paper and pen in hand
and an intuition to sharpen. People tend to put too much trust in simulations, and a lot of people get really ugly errors when using it from scratch.
It takes as much knowledge either way to design a circuit since you have to expect things.
Anyway. Clearly, we come to different conclusions, so one of us must be wrong. Statistically, note that I have a few people on my side...
Oh and BTW, I hope your neon light has an internal resistor.
Tim
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
About the dropping resistor for the neon , I never
saw a 120V rated panel indicator where they weren't
built in , with the dropping resistor already inside
the threaded housing . Sometimes I have used a meter on one at the rated 120 voltage to see what is the actual current draw for 120 , and learn what
the bulb itself is dropping , to calculate the added resistance if I put the same indicator across a 240 line or other voltage .
You wouldn't be nitpicking there , no.....
and yes I wondered for about two seconds
if I shouldn't show the dropping resistor for the panel lamp .....but you know I figured anybody skilled enough to
read the schematic already knows about neons , just
like that would be a fair assumption about the person
who wrote the schematic too , who is usually the same
one who designed the device .
The truth is I don't need the simulation for a sim I can run
in my head and see fine . I can read my own schematic
like I can read a shopping list for the grocery store ,
or the radio shack ......as the case may be , and I don't see any problem in this one so far as component function
and arrangement , all the signal paths and everything
flies the way it should ....if there's anything off it will
be fixed by a different capacitor or resistor value ,
and the ones I have now may be just fine .
So you have been looking at it with the old mental signal tracer , just like I have been doing , but you are seeing a problem somewhere.... so where
exactly do you see the snag ? I may be slow sometimes , but I'm not totally braindead so if you point me at the defect , I'll see it .
Where is it ?
About the SPICE , that was specifically regarding Twospoons virtual model earlier , and I suppose misdirected frustration ,
since at least you do tell me where you see a problem , even if it isn't a problem .....instead of saying just that my computer is smarter than your computer and my computer says it won't work ....just because it says so
.....but giving no further details
[Edited on 19-2-2006 by Rosco Bodine]
|
|
|
IrC
International Hazard
Posts: 2710
Registered: 7-3-2005
Location: Eureka
Member Is Offline
Mood: Discovering
|
|
The standard dropping resistor for neon bulbs is 150K to 180K 1/4 watt for 117VAC, but 1/2 watt would be better from a design standpoint. I think the
size is why they use 1/4 watt instead. For 240 VAC 300K to 320K would be fine.
[Edited on 19-2-2006 by IrC]
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
You know the really best looking panel lamps which
I have ever seen are made up custom jobs which
mount a high output narrow angle LED at a precise
distance behind a round fresnel lens where the
beam sort of " pops " in front like it was a virtual source
hanging in space a few mm in front of the surface !
I've never seen 'em sold , but I've made up a few of
them as a sort of " makers mark " indicators on custom equipment I have built .
After looking at the RC timing network and the signal level
which is available for switching Q4 , I don't think the NPN
is nearly as good a choice of switch as would be another
Mosfet . The base of the NPN is just trying to drink from
too small a stream of current there going to the timing capacitor and it will have marginal base current available
when the timing adjustment pot is set to provide the longest
duration startup pulse . The simple fix is just to put a Mosfet there to be Q4 and that's exactly what I'm going
to do to fix this .
Another possibility is putting a 555 there to do the timing ,
and get positive and predictable switching , whether the NPN or the Mosfet gets to be driven by the 555 . I like
555's , they are really versatile and perfect for scenarios like this one , so that's what is my inclination . Looks like this one
just won't be staying chip free , but a 555 is pretty basic .
[Edited on 19-2-2006 by Rosco Bodine]
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
On the 19th day of clicking keys and calculations ,
I think I just nailed it .
I was trying to get the job done using no chips
if possible , but every thing I looked at was
just inferior to using 555's for the timing ,
so now that problem is hopefully solved the simplest
way . Wish every time I run into needing a
non-inverted output I didn't have to scratch build
a timer using an op-amp , or use a dual 555 .
Anybody else like to see a 10 pin package evolution
version of the of the 555's , with dual outputs ,
inverting and non-inverting so you could use
which output you need .....or both ?
I believe I have this thing supplied in a way that
it will initialize and power up logic stable , in under
2 milliseconds , perhaps 3 at the most , with sufficent
reserve in the high voltage " virtual battery " capacitor
to maintain the 555's operation and the timing status
until steady state operation matures with the first complete
cycle . If I did my math right , no decimals out of place ,
it should work fine with the values I have assigned .
Anyway see if this revision looks better ,
because I do believe this one nails it .
|
|
|
Lambda
National Hazard
Posts: 566
Registered: 15-4-2005
Location: Netherlands
Member Is Offline
Mood: Euforic Online
|
|
Personally, I would go for an all Transistor/MOSFET approach seeing you put so much emphasis on rigidness and reliability. There are many Mono-stable
and Bi-stable Mulivibrator designs out there that use only 2 or 3 Transistors and Polystyrene or Mica Caps with very precise timing. If you however
insist on using Chips, why not go for the all in one Chip approach. There is a whole array of pre-designed Motor Control Chips, just waiting for this
job. But then you will be deviating from your previous design Philosophy, and Unique approach. Hell, you can even "Match" Transistors for minimal
temperature drift and precise timing applications. Keep it as simple as possible, then the least can go wrong. You have now just introduced more than
30 Transistors into you design. Unless you use MIL spec. 555's, you will be introducing temperature and reliability boundaries.
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
There used to be a dual CMOS version of the 555 ,
maybe LM7556 .....IIRC , but I haven't seen any being dumped as cheap surplus , or I would have used those
and not have to dump so many watts in the transformerless
power supply in the drop from 170 Volts to 10 Volts for
the timers .
I tried to tie into that supply in a way that wouldn't make the ripple worse than it is bound to be , figuring the 555's are stable enough to ride the
waves that get across the zener .
There's a lot of difference between 1/3 and 2/3 of the rails ,
so the huge hysteresis there should keep the logic state
latched , even with a crude and noisy supply like I have
improvised to keep from buying a 3 dollar transformer
But there's too much delay in the power up of a separate
DC supply , so I couldn't use that approach anyway . The
duration of the startup event is so short , and begins occurring at the instant power is applied , so unless the
mains to the motor were delayed until the logic stage was
powered up and the chips initialized , a conventional power
supply could not be used .
I looked at all the motor controller chips before beginning
this effort , since none of the motor chips have the capabilities of doing this from a purely hardware level .
You have to program an EEPROM to essentially be a BIOS
for your motor control system , and even then there's things
you can't do with the turn of a pot , that I can do here .
I have used 555's before in motor controls which are associated with refrigeration compressors , to sense
power interruptions as a " missing pulse detector " which
monitors the 60 Hz , and if the compressor is running when
any interruption occurs , the 555 assumes a compressor stall
before the event , and times the compressor out for two minutes , before allowing it to cycle again and see if the
current it is getting is within specs . So I don't see any
problem with using 555's in a motor control circuit , since
there already are numerous commercial devices which use the 555's in motor controls and they have no problems .
Edit above : LM7556 was the number for the dual CMOS
version of the 555 .
Anyway if it bothers you to think of a chip going into a
control system like this ......just try to think of a 555 as
a transistor with a few extra pins and a college education ,
a transistor array that will do what you wish a transistor
would do , but can't do that concert singing solo
[Edited on 20-2-2006 by Rosco Bodine]
|
|
|
Lambda
National Hazard
Posts: 566
Registered: 15-4-2005
Location: Netherlands
Member Is Offline
Mood: Euforic Online
|
|
OK, wants you are spinning it's apparently impossible to deviate you off course. The 556 is indeed the dual version of the 555 and can be purchased
via CONRAD Electronics International Mail Service for 1.51 Euro.
Conrad Electronics International Home:
http://www1.int.conrad.com
Search:
Electronics/& Metering --> Components ---> Active Components ---> IC's ---> IC's - Linear ---> Timer IC's ---> ICM 7556 IPD
In Audio Amplifiers it's fairly common to use power up delay circuitry which only turns on the final stage after the whole System has charged up
sufficiently, so nothing new here.
You can also use a Transformer with a primary of 110/220 volts input and a 110 Volts and 6-12 Volts secondary output. Now we are talking
International, with only one flip of a "Shift Switch" on the back side of the Magnetic Stirrer (50/60 Hz included).
I have uploaded ~100MB of Books and Motor Control Circuitry Designs to Madhatter's FTP services.
I cant Sing, so you will have to go on a Solo Tour by yourself.
[Edited on 20-2-2006 by Lambda]
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
| Quote: | Originally posted by Lambda
In Audio Amplifiers it's fairly common to use power up delay circuitry which only turns on the final stage after the whole System has charged up
sufficiently, so nothing new here. |
Yeah that's a nice feature in high power audio amps
having the filtered relays to eliminate speaker pops
that would appear during switching transients .
Of course it would be possible to get more elaborate
and go that route , but you couldn't hide the fraction of a second perceptible switching delay which would appear
at the toggle to the user , and the effect would be
" strange " to someone who doesn't know why they
have to stand there growing older by a hundred milliseconds
just bored and waiting for the stirrer to finally decide to
start turning A fast gun or a son of a gun could clear
a room full of trouble and be reloading in half that time
| Quote: |
You can also use a Transformer with a primary of 110/220 volts input and a 110 Volts and 6-12 Volts secondary output. Now we are talking
International, with only one flip of a "Shift Switch" on the back side of the Magnetic Stirrer (50/60 Hz included). |
Hey this circuit is Euro adaptable , just double the Ohm and Watts value for the the 3 power resistors below the bridge to allow for the increased
voltage , use a higher voltage capacitor . Also double the Ohm value or double the Watt rating of the 6.8K gate resistors for the output Mosfets , and
use 1 Watt pots on the main adjustments , and a higher voltage varistor for the higher voltage motor .......that's it .
| Quote: |
I have uploaded ~100MB of Books and Motor Control Circuitry Designs to Madhatter's FTP services. |
Yeah thanks for those . I have enough reading there to keep me busy for years
| Quote: |
I cant Sing, so you will have to go on a Solo Tour by yourself.
|
On some of the " industrial control " applications I like to
keep control circuits purely hardware logic where possible
and stick with linear chips . Everybody remembers the
" millenium bug " worries and what programmables might
be affected ......so there's no point in making basic control
circuits more complex than they need to be , just make
'em smart enough to do the job .....and too dumb to argue
It's the chauvinism of satisfied chip masters everywhere ,
enjoying the waltz of the flowers with satisfaction that these
obedient little minions know their place
[Edited on 20-2-2006 by Rosco Bodine]
|
|
|
Lambda
National Hazard
Posts: 566
Registered: 15-4-2005
Location: Netherlands
Member Is Offline
Mood: Euforic Online
|
|
OH man, ... Rosco !, ... you are just too bad, ... just too bad man.
How do I deal with this spinning wheel without cutting myself on those fine threads.
So now, ... you want to go IC, ... how about going PIC ? (don't read "Dick").
Eric's PIC Page:
http://www.brouhaha.com/~eric/pic/
Articles:
http://www.rentron.com/Myke3.htm
I have many Motor Control designs using PIC's, so just call the shots, and I will Shoot. You can tell (Program) this monkey to do anything, and if the
climb doesn't suit you, then just reprogram. You can preset a variety of programs, all to chose from, depending on what is desired. Maybe you want
your Hotplate to rise in temperature at 8 Degrees per 7 Minutes up to 73 Degrees, and gradually want the Stirrer to decline in speed from 600 rpm to
300 rpm during this period etc.
Now, I hope you won't get grey hairs over this one, ... will you. You can also start off by taking a few of mine, that will give you a "Head
Start".
Rosco !, ... you are just too bad, ... just too bad man.
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
LOL , I don't want an RS232 interface on my damn soldering iron
I want pure hardware logic all the way on this thing .
There was a news item today about a 62 year old
great grandmother having a baby .....
and I couldn't help thinking ,
" I wonder who's the lucky father ? " ....
I should be safe on that count , but there's always
the off chance of being the guilty party , you know
how these things go with paternity tests and all
|
|
|
Lambda
National Hazard
Posts: 566
Registered: 15-4-2005
Location: Netherlands
Member Is Offline
Mood: Euforic Online
|
|
Now, ... this is very nice of you, thank you for congratulating me on my new born baby.
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
My wife chuckled at the news and said later we
will probably hear that the woman was arrested
after it was learned the father was age sixteen ,
one of a procession of frequent visitors who
" came for milk and cookies " from the " nice old lady "
So how is it that a woman of her age becomes pregnant ,
well it must be like playing the piano , it just takes
practice .....practice ....practice , until you get there
The news did say she already had TWELVE children ,
so it would seem she has had plenty of practice
Anyway , not to get totally off-topic here ......there are
available TTL input points provided now with the circuit at its present evolution , the gate of Q4 being the most obvious
point where modulation from a feedback signal could be
applied to " servo-lock " a set speed which is tachometer
regulated . Any sort of programmed time and speed profile
which may be wished could be input to the gate of Q4 .
And there are other ways in which the circuit could be
accessorized with extended functions , but would not
be crippled with any of those options " NOT enabled " .
I wanted a circuit which would give good performance
inherently , good enough stand alone for just about anything , but also could be adapted to enhanced functions with later " add ons " , like
servo-locked tachometer based speed control or timed period sequenced speed profiles .
I would make those functions switched options on a
separate controller stage , a separate functional " block "
from the power stage I have shown here . Those options
and their schematic would be on page 2 , not yet written ,
and connected to page 1 by 3 or 4 wires if you follow my
meaning .
[Edited on 21-2-2006 by Rosco Bodine]
|
|
|
Twospoons
International Hazard
Posts: 1327
Registered: 26-7-2004
Location: Middle Earth
Member Is Offline
Mood: A trace of hope...
|
|
I preferred your turn on delay with just the cap and transistor (or you could use a fet).
And in defence of SPICE: its just a another tool in my design arsenal. Like any tool it has its place and its limitations.
Do you use a calculator, or do you work out square roots in your head? (I estimate in my head, then use a calculator - if there's a big discrepancy I
do it again!)
[Edited on 21-2-2006 by Twospoons]
Helicopter: "helico" -> spiral, "pter" -> with wings
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
There's a need for biasing a transistor which will
be used as a switch so that it has positive feedback
to the gate , so when it begins to switch it will make
a positive transition just like a relay , and hold that state
at a lower signal level than was required to transition in the first place , behaving a corresponding way when making the reverse transition . This
can be done using discrete components , on one transistor but better by using a complementary pair to create a bistable sort of
flip flop which toggles abrubtly to on or off and latches
there until a new transition signal level signal forces
a change of state . This is such a common task that
to save labor and the cost of using discrete components that chips like operational amplifiers and " timers " have been developed optimized for the
task . The label
of " timer " is actually a misnomer for the 555 type chip
since that timer function is only one of the applications
where it is defined in function by what external components are selected . I more usually use it as
a switching device having defined level transition
states at 1/3 and 2/3 of the supply rails , a signal
comparator which drives a flip flop controlled output .
It is a logic device arrangement of a very basic but useful form , a sort of rudimentary analog to digital converter . This takes all the uncertainty
out of the switching of
a single transistor which can fail to switch properly
under slowly changing or noisy signal conditions .
No chance of that happening with a flip flop , it has a binary output only , logic 0 or logic 1 state ......
fully off or fully on , and nothing in between .
I do use a calculator . The first scientific calculator I owned
was in about 1974 and it was a Texas Instruments Scientific with 1/2" Blue LED readout . It took four penlight batteries
to operate it , and IIRC in todays dollars it would have cost
over a thousand dollars at the time . But I often do the
familiar calculations in my head because I know the answer
in shorter time than it would take to key in the problem and
hit the equal key
[Edited on 21-2-2006 by Rosco Bodine]
|
|
|
| Pages:
1
..
5
6
7
8 |
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
