Emitter Follower + LPF Sine Convert for CD4060

Those with half a brain, and experienced with Laplace, will note that my analysis says "p" is the NORMALIZED Laplace variable.

If you're into "1 op-amp", you're already marked yourself as 1/4 brain, or LESS.

WHICH ARE YOU ?>:-}

Perhaps Larkinian-style ?;-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   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.

around a transistor rather than an op amp.

This is a 2nd order Sallen-Key lowpass driving an emitter follower. It's an edited version of John Larkin's schematic. The 10k at R4 gets rid of a nas ty kink in low point of the sine wave which appeared in the simplest revisi on.

I haven't played with the capacitor values at all - it would make sense to run the low pass filter at a higher impedance level (with smaller capacitor s) and buffer it's output with a second emitter follower running at a highe r current - perhaps a PNP device to roughly cancel the temperature dependen t offset. Version 4 SHEET 1 1132 680 WIRE 720 -96 48 -96 WIRE 48 -64 48 -96 WIRE 400 -48 368 -48 WIRE 896 -48 480 -48 WIRE 896 16 896 -48 WIRE 896 16 608 16 WIRE 608 64 608 16 WIRE -160 112 -256 112 WIRE -80 112 -160 112 WIRE 48 112 48 0 WIRE 48 112 0 112 WIRE 128 112 48 112 WIRE 304 112 208 112 WIRE 368 112 368 -48 WIRE 368 112 304 112 WIRE 544 112 368 112 WIRE 304 176 304 112 WIRE 608 176 608 160 WIRE 720 176 720 -96 WIRE 720 176 608 176 WIRE 1056 176 720 176 WIRE -256 192 -256 112 WIRE 608 208 608 176 WIRE 896 208 896 16 WIRE -256 352 -256 272 WIRE 304 352 304 240 WIRE 608 352 608 288 WIRE 896 352 896 288 FLAG 304 352 0 FLAG 608 352 0 FLAG -256 352 0 FLAG 896 352 0 FLAG 1056 176 OUT FLAG -160 112 IN SYMBOL res 16 96 R90 WINDOW 0 77 52 VBottom 2 WINDOW 3 88 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 1K SYMBOL cap 288 176 R0 WINDOW 0 64 30 Left 2 WINDOW 3 57 63 Left 2 SYMATTR InstName C1

SYMBOL npn 544 64 R0 WINDOW 0 101 17 Left 2 WINDOW 3 79 52 Left 2 SYMATTR InstName Q1 SYMATTR Value BC547C SYMBOL res 224 96 R90 WINDOW 0 79 43 VBottom 2 WINDOW 3 -48 53 VTop 2 SYMATTR InstName R2 SYMATTR Value 1k SYMBOL voltage -256 176 R0 WINDOW 0 28 122 Left 2 WINDOW 3 11 230 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value PULSE(0 6 1m 1u 1u 30m 60m 100) SYMBOL voltage 896 192 R0 WINDOW 0 -70 93 Left 2 WINDOW 3 -61 132 Left 2 SYMATTR InstName V2 SYMATTR Value 6 SYMBOL res 592 192 R0 WINDOW 0 -69 39 Left 2 WINDOW 3 -75 76 Left 2 SYMATTR InstName R3 SYMATTR Value 100 SYMBOL cap 32 -64 R0 WINDOW 0 64 30 Left 2 WINDOW 3 57 63 Left 2 SYMATTR InstName C2

SYMBOL res 496 -64 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R4 SYMATTR Value 10k TEXT -120 328 Left 2 !.tran 1 TEXT -224 40 Left 2 ;SORTA SINE WAVE That's a high beta NPN transistor BC547c. Save that text as Sorta_Sine_2.asc and open that in LT Spice. Click the run icon and probe.

--
Bill Sloman, Sydney

You could do a mosfet version too, if a trapezoid is close enough. Tweak the cap to tune the slopes. It swings the full supply voltage, and lightly loads the cmos gate.

Version 4 SHEET 1 904 680 WIRE 448 16 368 16 WIRE 608 16 528 16 WIRE 800 16 608 16 WIRE 608 64 608 16 WIRE 32 80 -64 80 WIRE 112 80 32 80 WIRE 368 80 368 16 WIRE 368 80 192 80 WIRE 560 80 368 80 WIRE -64 160 -64 80 WIRE 800 160 800 16 WIRE 368 176 368 80 WIRE 400 176 368 176 WIRE 608 176 608 160 WIRE 608 176 464 176 WIRE 672 176 608 176 WIRE 608 208 608 176 WIRE -64 320 -64 240 WIRE 608 352 608 288 WIRE 800 352 800 240 FLAG 608 352 0 FLAG -64 320 0 FLAG 800 352 0 FLAG 672 176 OUT FLAG 32 80 IN SYMBOL res 208 64 R90 WINDOW 0 77 52 VBottom 2 WINDOW 3 81 51 VTop 2 SYMATTR InstName R1 SYMATTR Value 10K SYMBOL cap 464 160 R90 WINDOW 0 75 31 VBottom 2 WINDOW 3 81 31 VTop 2 SYMATTR InstName C1

SYMBOL res 544 0 R90 WINDOW 0 -56 51 VBottom 2 WINDOW 3 -48 53 VTop 2 SYMATTR InstName R2 SYMATTR Value 7K SYMBOL voltage -64 144 R0 WINDOW 0 28 122 Left 2 WINDOW 3 11 230 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value PULSE(0 6 1m 1u 1u 30m 60m 100) SYMBOL voltage 800 144 R0 WINDOW 0 -84 92 Left 2 WINDOW 3 -72 123 Left 2 SYMATTR InstName V2 SYMATTR Value 6 SYMBOL res 592 192 R0 WINDOW 0 -71 43 Left 2 WINDOW 3 -75 76 Left 2 SYMATTR InstName R3 SYMATTR Value 100 SYMBOL pmos 560 160 M180 WINDOW 0 86 73 Left 2 WINDOW 3 64 36 Left 2 SYMATTR InstName M1 SYMATTR Value ZXM62P02E6 TEXT 72 296 Left 2 !.tran 1 TEXT -80 8 Left 2 ;SORTA SINE WAVE, PFET VERSION

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation

The schematic simulates amazingly slowly for such a simple circuit - 55nsec per second, or 5hours (on my computer) to run a full second of simulation.

The Zetex ZXM62P02E6 does have a bit more bandwidth than this circuit calls for.

and it isn't exactly tightly specified.

--
Bill Sloman, Sydney

e

-variable filter for a Butterworth when you can use 1 op-amp either. Looks like its been designed by a physicist or some non engineer with a basic ele ctronics knowledge.

Don't add to the ignorance. No engineer uses p! I don't care if it is sCR o r not it is non standard and daft. Also that state filter is not really a s tate filter. It is just two first orders in cascade with feedback. A proper state feedback filter has two integrators with feedback of a defined amoun t as per a signal flow graph. It is a filter but not a standard one and poi ntless. Looks like its been designed by a physicist or student.

Thanks again. I will check out all the variations you have suggested on the bench tomorrow.

Martin King

Thanks for your contribution Bill. Between this, and all the other suggestions made so far, I am sure I can now come up with an acceptable circuit.

Martin King

Matching the ability to position the poles accurately against available components; I used to make 5-pole LPF's using 2N3906 followed by the next stage 2N3904 so the DC offset was minimized.

If you do use single transistors like a 2N3904, or 2N3906, be sure to compensate your Sallen-Key equations.

Afraid a simple adjustment of the gain to 0.99, or 0.992, or 0.995 and adding a tiny bit of output resistance won't do it either. Seems there's a 'substantial' feedback term in there that is neglected with those simplifications.

I know the old MultiSim PSpice will show you the difference in the response curves. Probably LTspice will too.

Try a side-by-side filter analysis using transistors and using simple 0.99 controlled voltage source blocks. You'll be surprised.

If you were happy with 1 pole from John Larkin, and even liked better the

2 pole from Bill Sloman, why not use BS' 2 pole and add an R and C at its input and make a standard 3 pole LPF? with the single pole at the input you can make the pole pair a bit more 'ringy' and get faster roll-off at the higher freq's

Also, if you have trouble driving ANY of the terminals on those output components to outside the rails in order to get the drive to take the output TO the rails; you could add a bit of 'charge pumping' to take the drive higher [or lower, depending]. In otherwords take advanatage of the fact that a KNOWN constant AC signal is coming into the LPF.

An NPN that's been around for awhile beta of over 600, peaks at low current, lower than where 2N3904 peaks. These used to be used in those cheap telephone line HOLD/MONITOR circuits, because they could monitor the telco lines and only 'eat' a micro of current, pass the then FCC Part 68, [is that the right number?] Haven't seen the 2N5088 anymore in the TO92 package, since everybody went to SMT.

watch for wrap: .model Q2N5088 NPN(Is=5.911f Xti=3 Eg=1.11 Vaf=62.37 Bf=1.122K Ne=1.394

+ Ise=5.911f Ikf=14.92m Xtb=1.5 Br=1.271 Nc=2 Isc=0 Ikr=0 Rc=1.61 + Cjc=4.017p Mjc=.3174 Vjc=.75 Fc=.5 Cje=4.973p Mje=.4146 Vje=.75 + Tr=4.673n Tf=821.7p Itf=.35 Vtf=4 Xtf=7 Rb=10) * National pid=07 case=TO92 * 88-09-07 bam creation

Ignorance in abundance! The only Churchill you know is the one who was Prime Minister.

Normalized solutions are common.... then easily scaled to the desired frequency.

If you'd learn to read, my configuration is a twist on state-variable that eliminates dissipation factor effects in filters used at low audio frequencies.

Build a standard state-variable filter at 16Hz and you'll _measure_ the problem... something mathematical analysis or flow graphs will not.

I'd suggest you run the math, but it's clear you can't.

So maybe simulate it, if you can even do that?

Aha!! I see the problem... you are P-brained >:-} ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   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.

MultiSim =/= PSpice

...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   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.

That guy finally did it for me. I've reinstated the "Trash Google Posted" filter that I removed when things got better, a while back, for a bit.

I just hope my whitelist is up to date.

--
"Design is the reverse of analysis" 
                   (R.D. Middlebrook)

8038 & Max 038.

--
Anyone wanting to run for any political office in the US should have to 
have a DD214, and a honorable discharge.

The sad part is that " snipped-for-privacy@gmail.com" is probably a PhD candidate... he has all the earmarks... profoundly ignorant and inexperienced, but smugly insistent >:-}

I currently have 132 individuals white-listed. I watch threads for newcomers who make cogent comments and add them to my white-list. There are even some gmail/google types in my white-list >:-}... they aren't ALL bad, but I do wish they'd use something other than googlegroups to post, the excessive white-space is a real annoyance. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   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.

Or more, if you're into substantial range plus very low distortion...

OpAmps make it much easier, because you can make precise clipping points. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   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.

People's choice of user name can give valuable clues. I try to avoid anyone with a sci-fi-sounding, or pretentious 'nym.

Having said that, that particular name may be genuine, Czech, Slovak, Serbo-Croat, or summink ;-)

I'd have thought that it wasn't beyond the wit of Google's programmers to write a proper, threading, news server/client. Form over function, I guess.

--
"Design is the reverse of analysis" 
                   (R.D. Middlebrook)

We used to use a ramp into an OpAmp with a series of transistor followers in the feedback with each transistor base connected along a voltage ladder. 'Squashed' the signal fairly rapidly and was extremely easy to calculate. Required less than four stages to get really good waveforms due to the 'soft' turn on.

The OP said that a rounded square wave would be adequate. And he's driving LEDs from a voltage source, so can expect a lot of distortion on the low end.

And he said that the RC+emitter follower was good enough.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation

Going directly from square to sine is very challenging. Most people convert square to triangle, then use transistors to round the wave into nearly sine. That is how it was done in the Intersil 8028 and the Exar

2206.

It takes about 8 to 12 transistors, both npn and pnp.

?-)