Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently Jun Mon 18, 2018 10:34 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 93 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next
Author Message
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Mon 05, 2018 5:45 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3453
Also instead of buffering with an opamp, it could be done with a Schmitt trigger IC to maintain the quick transitions.

Here's an excellent discussion about discrete voltage followers: http://sound.whsites.net/articles/followers.html#s2


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Mon 05, 2018 6:59 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
Ooops I wrote "Schottky" but I meant "Schmitt trigger". I have a bunch of 7404 hex inverters on hand, used two inverters to make a Schmidt trigger. Truth is, it worked fine when I buffered with a single inverter but I figure the snap action of a Schmitt is more bullet-proof.

The op-amp is what is giving me a bunch of gain from the generator.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Mon 05, 2018 8:21 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Here is what I am seeing. Scope is at 2v/div. At 9 volts the unloaded wave form is about at the rail voltage but is not square. Inserted into the circuit it drops to about 5 volts and the dip in the middle of the top varies a lot with the antenna tuning. Looks like Q1 is not really unhappy but it may be loading the PLL. Running at 12 volts the unloaded waveform does not change it just increases to near 12 volts. So supply voltage does not appear to be an issue.

I am getting a shopping list together to order the synthesizer parts but also planning on ordering a few CD40106 CMOS Schmitt trigger. Any suggestions on P channel FETs? I have none on hand. I think that may be an improvement to the ECS version as well.

Also the ringing we saw with the ECS on the RF waveform is barely detectable with the PLL. But that may be due to the slower rise time. Not sure.


Attachments:
20180305_124244_resized.jpg
20180305_124244_resized.jpg [ 55.84 KiB | Viewed 546 times ]
20180305_124334_resized.jpg
20180305_124334_resized.jpg [ 63.88 KiB | Viewed 546 times ]

_________________
Rick
AB5ON

Forest Gumping through life
Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Tue 06, 2018 2:51 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
That PPL chip is a lot better at sinking current than sourcing it, according to the datasheet, and your bendy 2nd trace is a classic example. I think faster rise and fall times time will increase your radiated output too. It would be interesting what effect a 2k pullup resistor would have, from synthesizer output to 9v supply. A Schmitt trigger buffer might help BUT you'll still want that pullup resistor to make sure the square wave duty cycle (as sensed by the trigger input) remains very close to 50%. BUT, the output of a CD40106 is only a little better at sourcing current than the PPL chip, I think there would still be a slow rise time at its output, so I'm not sure you're gaining much. Try a pullup resistor and see what happens.

In my tests here, our single transistor modulator performs best when the base is driven by a low impedance square wave with fast rise and fall. Q1 seems happiest when it can pull a few ma of current from the square wave source when the square goes high, which is exactly where your slow rise time needs to be fast. HCT logic does a great job at sourcing this and seems equal to the ECS' output in that they both source current high and low. So, one solution that is easy and should work is to add an HCT type of gate to buffer the PLL's output. I am using a 74HCT04 with a couple of resistors to form a Schmitt trigger, which works brilliantly and I still have gates left over, but you could use a gate designed with that type of input. Either way, might not need a dedicated 5v supply, I haven't tried this, but a simple resistor to drop a few volts with a 470u or 1000u cap across the gate's supply pins (to give the chip a reasonably steady supply) could do the job. May not even need to worry about logic level difference from 9 volt cmos to TTL, the guerrilla way is just hook 'em together and pray. Maybe one of the buffers Mac has linked to above can be pressed into service instead. Problem will be to make the logic low voltage as low as possible, below .7 volts. Or, rework the whole shebang for bipolar supply.

This is what comes out of my reshaped rf generator, from an HCT gate, 2 volts/division. This is operating in circuit and the little wrinkles on logic high is caused by Q1.
Attachment:
1mHz ttl buf out.jpg
1mHz ttl buf out.jpg [ 51.76 KiB | Viewed 537 times ]


The above trace is taken at the oscillator side of R1. Following is the Q1 side. R1 is 500 ohms, Q1 base is drawing about 5ma from the gate during logic highs
Attachment:
Q1 base 5ma draw.jpg
Q1 base 5ma draw.jpg [ 46.79 KiB | Viewed 536 times ]


This stuff gets complicated doesn't it? I have another post coming about keeping audio distortion as low as possible with this circuit. Probably tomorrow.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Tue 06, 2018 5:01 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
richfair wrote:
Try a pullup resistor and see what happens.

So, one solution that is easy and should work is to add an HCT type of gate to buffer the PLL's output. I am using a 74HCT04 with a couple of resistors to form a Schmitt trigger,


Bingo! Rich you were right on the money. A pull up resistor is the simple fix. I hooked up a 5k pot and adjusted until the dip just went away. Measured at 3.7k. Went back and tuned the antenna and only the tiny bit of ringing at the top moved. I may still try out your approach to the Schmitt trigger to compare them. We may be close here. The issue I was seeing with the distorted output turned out to be my antenna. I was using a really long wire in my attic I use mostly for receiving. With a shorter antenna in my office I did not see the problem.

I like your project using the signal generator as the input to the transmitter final. It is nice to adapt what you have on hand. I experimented a bit by just using a single transistor as an RF amp. Ran my iPhone into the external audio of the signal generator. Had some success with it.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Tue 06, 2018 2:26 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
Wow! That's fantastic news, so good when something works the way it should! "That tiny bit of ringing at the top". Would you tell me, again please, at exactly what point you are viewing that on?

This rf stuff seems to show up everywhere, getting into places I don't want it and causing trouble. So you've been using a long wire in the attic, and were able to match to it okay? Interesting!

One thing I'd suggest, that I am finding helpful, is to replace R1 (Q1 base resistor) with a 1k trimmer. I've found that R1 can be adjusted to optimum audio harmonic distortion, a different value for different Q1 devices. I don't know why exactly. I'll include this in a later post about audio distortion but figured its worth mentioning now in case you're putting in a parts order today.

I'm going to build one of Mac's frequency sythesizers once I am able.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Tue 06, 2018 2:47 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
richfair wrote:
"That tiny bit of ringing at the top". Would you tell me, again please, at exactly what point you are viewing that on?

replace R1 (Q1 base resistor) with a 1k trimmer. I've found that R1 can be adjusted to optimum audio harmonic distortion, a different value for different Q1 devices.


Look at my closeup picture of the scope near the top of page 2. In fact on a number of our square waves at the top of the positive leading edge there is a slight decaying ring. At the point I took the picture we were still working on the resistors, transistors, diode, etc. I think we got it under control. But the PLL seems less prone to it. I still see just a bit of it on the PLL output. The pull up resistor gets rid of the dip then reduces the ringing but the output then suffers. Does not seem to be an issue so I am OK with it now.

Good idea for R1. I need some 1k trimmers anyway. How are you determining audio harmonic distortion?

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Tue 06, 2018 3:47 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
I've been a little confused about one thing.
Quote:
Look at my closeup picture of the scope near the top of page 2.
That trace is obviously not from the ECS output. Is it audio or rf (timescale)? WHERE was your scope probe attached? More importantly, is it still present?

Quote:
In fact on a number of our square waves at the top of the positive leading edge there is a slight decaying ring
I agree that some minor ringing won't hurt anything, as long as the ripples aren't strong enough to cause Q1 to be somewhere in its linear area between full on/off. On my bench here, a 6 inch piece of wire between the oscillator's output and R1 is enough to make the trace at R1 give you seasickness. Its a reminder to keep all wires as short as possible!

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Tue 06, 2018 4:02 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
Quote:
How are you determining audio harmonic distortion?
Ah! Another guerrilla tactic. I am using a portable radio that has little or no bass output. I broadcast a 100hz sinetone (at high-enough modulation level for harmonics become obvious, around 90% or so) and tune in on the little radio. It doesn't reproduce 100hz but the 2nd, 3rd, 4th harmonics ARE easily heard. I know I am not listening to harmonics generated by the radio itself because the harmonics are totally dependent on modulation level, not on listening volume or signal level to the radio (I can walk away from the antenna with no change). Adjust R1 to minimize harmonics. Antenna peaking may factor into this since harmonics become stronger with higher current through Q1.

One of the things I want to do today is take a closer look at the harmonics on an RTA and see if there is anything further we can do to minimize them. The last thing I want to do is complicate this circuit any further, simple is much more interesting, and it works really well already.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Tue 06, 2018 11:06 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
richfair wrote:
I've been a little confused about one thing.
Quote:
Look at my closeup picture of the scope near the top of page 2.
That trace is obviously not from the ECS output. Is it audio or rf (timescale)? WHERE was your scope probe attached? More importantly, is it still present?


That waveform is the RF output at the inductor with my scope probe loosely coupled to it. It it now barely detectable. I will post some pictures a little later.

Love the guerrilla tactic. Nice way to check. I have a small transistor radio that I use for testing. Since they are easily moved about and reoriented they are useful for other tests as well.

Had a little issue with RF into the audio this morning that had not been there before. Only to find out my antenna wire was laying across the charging cable of the iPhone I am using for audio input. DOH!

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Wed 07, 2018 12:15 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Waveforms with top trace showing the PLL output and bottom the RF output at the antenna. Voltage at 9 volts. Scope at 1 volt/div.


Attachments:
File comment: No pull up resistor
wave1.jpg
wave1.jpg [ 66.57 KiB | Viewed 490 times ]
File comment: Pull up resistor set to just pull some of the dip out of the top. Output decreased some.
wave2.jpg
wave2.jpg [ 73.83 KiB | Viewed 490 times ]
File comment: Pull up resistor decreased until waveform was clean but output is almost gone.
wave3.jpg
wave3.jpg [ 63.2 KiB | Viewed 490 times ]

_________________
Rick
AB5ON

Forest Gumping through life
Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Wed 07, 2018 2:38 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
Very interesting! So a pullup resistor, of any value, does nothing but decrease radiated output? Or is there a sweet spot?

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Wed 07, 2018 3:36 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
richfair wrote:
Very interesting! So a pullup resistor, of any value, does nothing but decrease radiated output? Or is there a sweet spot?


The pull up changes the shape of the PLL output a lot. There is sort of a sweet spot but it presents a problem also. The sweet spot is under 40 ma of current. I have a 100 ohm resistor in place of R2 so I can test a range of output levels. At higher levels the RF signal is distorted and by 90 ma it is not useable. So maybe not as simple as we hoped but learned a little more about the PLL output. Not a complete failure. Additional testing found that the dip in the PLL output can actually be reduced a lot by antenna tuning. If you peak the antenna and then reduce capacitance from peak, the dip in the waveform is reduced to a reasonable level without losing much output. That is good news but I think just a temporary work-around since you need to tune with an oscope.

Yesterday was a pain. Spent several hours chasing RF in the audio, distortion and poor modulation. Took the transmitter back to a known good configuration with the ECS and still had the problem. Finally discovered the iPhone I was using had failed and was causing all the issues. Put another spare iPhone on and it worked just fine. :evil: Pretty much shot the morning. Did have an interesting experience working with the transmitter and no radios on and just watching the scope. I started hearing a tone that matched the audio input frequency and volume, but I had no receivers on. Tracked it down to an old pay phone I have in my office. The RF was getting into the phone and coming out the carbon element even with the phone on the hook!. It may have been doing that all along but with a radio on I didn't notice.

I would still like to order a P channel FET to test. Have any suggestions? I would like to include that in my parts order.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Wed 07, 2018 5:55 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
Sorry to hear about yesterday's rf extravaganza on your bench! I'm working up a much longer post but want to get a reply in here now, since we're in the middle of another "snowmageddon" this morning. Power may disappear at any moment.

I've heard my output coil (wire around a large pill bottle) singing with audio, but having it come out of a phone's carbon transducer is really unique. I don't have a ready p-channel suggestions. Somewhere out on the web, maybe linked by Macrohenry in another thread, using the same approach as ours but with a mosfet instead of an npn transistor. That would be worth playing with.

I have to go help a neighbor now...

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Wed 07, 2018 9:27 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
Neighbor is okay and so is our power. So ........

Without further ado,

Rick, I’m thinking I may have gone about a far as I can. The transmitter sounds very good as it is, strong output and clear sound with very few components. I worry that if I write more about its weaknesses, doing so will turn people away from this design. They should try it! I am extremely pleased with mine.

We've learned that the oscillator needs to produce a sharp 50/50 squarewave for best output. The ESC oscillator works brilliantly although it is not guaranteed for better than 45/55 symmetry. Less symmetrical squares can reduce radiated output and add spurious junk. With some experimentation, R1’s value can be tweaked for improved audio performance as well as reducing wasted current draw through Q1’s base. Q1 adds a bit of audio harmonic distortion when the audio level rises. It is minute and mostly inaudible until around 75% modulation and above. Above 75% is the upper 3 db of dynamic range, which means the distortion is primarily only on transient audio peaks and as such, probably inaudible. Being an audio guy as I am, it doesn’t bother me. Since R1 can also affect modulation depth the best setting is not always easy to arrive at. 200 ohms seems to be a good value for those who don’t feel like tweaking.

On my unmodulated radiated rf output I can see traces of unwanted ringing, small ripples on the upper/lower points of the waveform. The ripples frequency is higher than the tuned frequency and seems unrelated to the tuned frequency. I can feed a clean square wave directly to L1 and see the ripples on the radiated output, so I am sure they are not introduced by Q1. I don’t hear them on a radio and they are generally far lower in level than the fundamental, so probably harmless.

C4 cap from Q1 collector to ground is another thing to be experimented with. I don’t want any cap there but it can help control those weird ripples that show up in the radiated signal. My guess is that by smoothing off waveform peaks of Q1’s output, less energy excites the "thing" that causes those ripples. What that "thing" is, I don't know. Maybe reflections within the antenna? Anyway, C4 also can reduce some instability I've seen on modulation peaks when tuned to certain frequencies, in my case around 1500kHz. I did not determine the cause but C4 usually fixed them with no downside I noticed.

My best foot forward design:
Attachment:
Modified 35Z5 transmitter rev3 sml.jpg
Modified 35Z5 transmitter rev3 sml.jpg [ 95.86 KiB | Viewed 462 times ]


edit: OMG I just started looking for suitable, sourceable variable caps. I've been out of touch with this corner of electronics! A 300v or greater part is called for but new choices are expensive and all of them are fiddly teeny blade trimmers. Looks like it is time to steal something off of an AA5 carcass or visit an auction site. My flimsy little 100v 3-40pf part isn't going to cut it, in fact maybe that is why I'm having problems with high modulation peaks, I'm sure the peaks are way over 100v.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 08, 2018 3:23 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Rich;

Thanks for all your help! You made a huge contribution to this project. We started with an unstable circuit that drew as much as 250ma and got the components hot to a stable design that can accommodate a variety of transistors and has good modulation at full legal power. The other feature I like is adjusting R2 to change the output power without throwing everything else out of whack. Thanks for figuring out the inductor and resistor to make that possible. This approach also opens up the possibility of using another oscillator similar to the ECS but that cannot be modulated. Other frequencies may now be possible with this design. With the specs for output you discovered others can look for alternate devices that should work.

I think I will continue to work with the frequency synthesizer. Your idea for the Schmitt trigger buffer needs to be checked out. For now I have found the settings that work with the synthesizer so it is a working frequency agile transmitter. Just a little picky to set up. The buffer may solve that.

Need to acknowledge Tom (35Z5) for his work on the original ECS oscillator and the antenna matching design. That was the foundation for this entire effort!

This has been a fun project and enjoyable collaboration.

Regards,

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 08, 2018 4:13 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12567
Thanks guys you've put more effort into this Tx than I did the first... No doubt two heads are better than one...

I'm going to dub it Rick & Rich

You guys send me a PM with addy & I'll send you each a Dale 1500Kc osc.. They Modulate poorly in my design but should be fine in the Rick & Rich, I've used them as driver in tube Tx...

_________________
Tom


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 08, 2018 5:38 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
Tom, thank you that is very generous of you. This thing works with different frequencies as long as the matching network is adjusted. I know that you and other forum gurus know how it all works, to me it is still a mystery. Much to learn about there!

Rick, we're agreed that a frequency synthesizer is the best path from here. I think you'll benefit greatly from reshaping your (Collin's? Mac's?) synth to speed up its output transitions. Slow transitions will prevent Q1 from cleanly switching on/off and most probably will change its on/off duty cycle further away from 50-50. Hopefully reshaping with a Schmitt trigger will not mess up symmetry too much. On my bench it is easy to see the bad effects of non-symmetry. I can drive Q1 from a reshaped Heath oscillator output, using either a near-perfect 50-50 square or a pulse wave having a duty cycle that ranges from 38-62 to roughly 10-90. The electrical characteristics of those outputs are otherwise identical. Transmitter's radiated output falls and audio harmonic distortion rises as the symmetry gets further away from 50-50. There is no doubt in my mind that a clean and sharp square wave for R1 is required.

Thanks again for letting me play with this. I've learned a great deal and I look forward to where you'll take the "Rick&Rich" next!

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 15, 2018 5:03 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Update on frequency synthesizer. I got in the 7404 chips and made a Schmitt trigger from them but also ordered 7414 Schmitter trigger inverters. Even tried just the 7404 as a regular inverter/buffer. They all clean up the square wave just as they should. However the output is not as close to 50/50 duty cycle as I would like. The Schmitt trigger was actually the worse for symmetry. Revisited the pull up resistor using a pot to vary the value. Did not help at any setting. Just out of curiosity I tried it as a pull down and actually got some improvement but as it approached 50/50 it gets really touchy and the frequency gets off.

I did notice on the frequency synthesizer Colin built and sent me for testing there was a .01 uf capacitor which is not in the schematic on Macrohenry's site. I tried it without the cap and the transmitter would not work. Put the cap back in and it was fine. Then I put the ECS back in and tried it with the same value cap and it worked just as well either way. With the cap in place and no buffer the transmitter works about the same as the ECS when voltage is around 9 volts. You can get a little more output out of the frequency synthesizer with a higher voltage but not much. At 9 volts the ECS output is right at 5v p/p. At 9 volts the synthesizer output is just over 6 volts when loaded by the circuit. Dropping the voltage I found the synthesizer is stable down to about a supply of 7 volts before it complains. So at 9 volts which we had found was about right for this design you can pretty much swap the ECS and synthesizer at will. Modulation on both looks to be the same with just over 90% pretty clean.

I ordered parts for the synthesizer to built another one and found the Motorola PLL is now obsolete with some NOS on eBay and other surplus suppliers. Checked out some alternates and ended up ordering a TI CD4046BE which dropped right in and seems to work fine. So when the supply of Motorola chips dries up there is still a source that will work. I will get my order list typed up since there are a couple of part number changes that have occurred since Macrohenry designed the circuit.

Tom is sending some of the oscillator chips that cannot be modulated for testing in this circuit. I will post the results as soon as I test them. There may be some further refinement to this possible but well beyond my skill level. It is working just as I hoped when we started and certainly better than I could have done alone. Now to build a final version rather than the proto-board set up I am using. If anyone else has any suggestions or ideas feel free to jump right in.

Rick

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 15, 2018 6:11 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 510
Location: Equinunk PA 18417
It is true that a Schmitt trigger will alter the duty cycle, an unfortunate consequence of the way it achieves sharp transitions with less-than-perfect inputs. A good way to get a 50/50 square is to double the oscillator frequency, then divide by 2 using either a counter or a D-flipflop. Either will make a near-perfect square. Here is the shaper I am using with a Heath rf gen. (Please no comments from the gallery about what a sloppy imperfect circuit this is. I'm running the opamp open loop, down and dirty but it works.) To get a 1mHz 50/50 square, the gen is set at 2mHz, the resulting pluses are detected by the 74hct74 which outputs a very nice 1mHz square.
Attachment:
File comment: 5v supply connections to TTL chips are not shown. 5 volts comes from the 78L05 of course.
rf generator to ttl squares.JPG
rf generator to ttl squares.JPG [ 97.42 KiB | Viewed 308 times ]


My plan for a synthesizer has been to double the PLL's frequency output and use 20kHz steps, running that output through a flipflop for the desired frequency with 50/50 squares. I haven't explored the idea deeply though I am sure the changes are simple. For instance it looks like you can move the connection from pin 15 of the 4060 osc/divider to pin 13, to get a 20kHz output. I'm not sure what changes need to be made around the 4046 PLL to double its frequency. 9 volts maybe high enough for the PLL to run twice as fast, I don't know, maybe not.

_________________
-Richard Fairbanks


Top
 Profile  
 
Post New Topic Post Reply  [ 93 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: Google [Bot] and 3 guests



Search for:
Jump to:  
























Privacy Policy :: Powered by phpBB