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


It is currently Jun Fri 22, 2018 12:09 pm


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: Feb Wed 28, 2018 6:53 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12578
richfair wrote:
Tom, the key is your tank matching circuit. It really builds up a head of steam.

I'm now remembering the first time it tuned up, had to connect the scopes 10x probe to be able to read it... Couldn't believe what I was seeing...

_________________
Tom


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 01, 2018 12:18 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Some additional notes on voltage. I put in the LED and adjusted current back to 90ma. The voltage to the 386 measured just above 5 volts so I pulled the LED and moved the power for the 386 over to the 5 volts from the 7805. At 12 ohms for R2 it was pretty much at 90ma which I think is about the optimal for this circuit. Then I put my 75 ohm pot back in for R2 and started dropping the output while listening to my Philco 48-482 on the table next to me. Took it down until there was noise and then back up until the hiss went away. That was at about 30 ma. Ran some calibrated ear audio checks :-) and the audio was really good there and not as good at full power. I suspected this but had no real way to try this until now. With the output power able to be controlled by R2 you can run this much lower for radios in the same room or crank it to full output for the radios on the other end of the house. Setting R2 at 30 ohms puts the low output where I want so I may just wire in a HI/LO switch.

With the 5 volts able to produce the target of 90ma I don't see a need for the 9 volts to be applied to the 386. This should help the 386 also by having a regulator right on the board near it.

I still have not tried other transistors but believe setting the value of R2 for 90 ma may account for variation in transistor properties. This may be part of the final set up when building.

Enjoying some big band right now.

Edit; when moving power leads around to test DO NOT apply 9 volts to the ECS oscillator. It won't like that. Don't ask how I know this. :-)

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 01, 2018 1:24 am 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1368
Location: Mount Pleasant UT 84647 (from Coventry, UK)
black85vette wrote:
. Now that it is more stable I think I would like to experiment with other oscillator sources including Macrohenry's synthesiser.

Rick

I would love to see the results of that experiment, as I have one of those synthesizers sitting around doing nothing. I have enjoyed following this thread so far.

Colin


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 01, 2018 3:06 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 513
Location: Equinunk PA 18417
Yeah the ECS oscillators do not handle overloads gracefully. I lost one too. Over voltage or an accidental short of its output to ground will kill it. Regarding 5v supply to the LM386, I agree with you that the circuit performs quite well that way. I also agree you may be able to skip a bypass cap on the LM386 that way, with good layout. A 5v supply and an R2 of just a couple of ohms (or even eliminated) still gives a power level at or above legal limit with clear sound.

Most of my antique radios, those I did not sell or give away last spring, are still in storage but I have 4 portable radios around. Two of them do not handle strong signals well and sound different shades of awful when in the same room as this transmitter (unless I de-tune the antenna to reduce rf radiation). The other two sound good all the time. Is it possible you are experiencing a similar problem in that your radio is simply unhappy with a strong signal?

Some folks would prefer a 9 v supply, some a 5 volt, some perhaps a 12-13 volt supply. This circuit can be easily adapted just by adjusting R2 (and of course making sure the ESC receives only 5 volts!). That said, I am trying to run down a distortion issue. There may be two separate issues actually. I am seeing harmonic distortion components introduced that are a little different for each batch of Q1 devices. In general, all devices work pretty well until modulation is pushed up past 80% or so, then the harmonics start to creep up until they jump dramatically just before 100%. Thing is, they seem to jump up less when I operate with 9 volts. To be honest, I don’t hear distortion at all with program and I don’t claim to be “golden eared”. I need another day to sus it out better. This design forces the LM386 to operate very near its maximum output before clipping, which is true whether using 5 volts or greater supply. I took a step today to fix that. By paralleling a big 470uf cap across R2, the AC (audio) drive is raised as high as possible, independent of the DC level imposed by R2. Adjusting R2 no longer also adjusts the relative modulation level. This way, with reduced power output (to legal limit) the LM386 does not have to work as hard and therefore may give lower distortion. Yet, after doing this mod the harmonic distortion near max modulation remains. I’ve noticed that the harmonic components can be changed, and reduced, by detuning the antenna match a bit. With 2n4401 I can get harmonics lower than with other devices just by adding on a couple additional pf after finding the antenna peak. More work to be done in that area. It may be unavoidable with a single transistor modulator. I’m actually impressed that it works as well as it does!

Colin, thank you for your support! I started working on an interface for my old Heatkit rf generator. Can you give me a link to Mac's synthesizer? Oh, Rick, I don't know why you haven't received my mail yet with some other devices. Maybe tomorrow.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 01, 2018 3:21 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Yes, too much power too close is exactly what I experienced. Increasing R2 now cures that.

I think I have seen the same harmonics you noticed. They did not appear to affect the audio that I could hear. They do vary with the tuning of the antenna. I did not make a serious attempt to eliminate them. You can actually hear it with a sine wave and increasing the modulation to near 100% The tone changes slightly. I think it is happening near the peak of the RF carrier. It has ringing that moves up and down the positive slope when tuning the antenna.


Attachments:
rfwaveform.jpg
rfwaveform.jpg [ 37.69 KiB | Viewed 582 times ]

_________________
Rick
AB5ON

Forest Gumping through life
Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 01, 2018 5:18 am 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1368
Location: Mount Pleasant UT 84647 (from Coventry, UK)
richfair wrote:

Colin, thank you for your support! I started working on an interface for my old Heatkit rf generator. Can you give me a link to Mac's synthesizer?

viewtopic.php?f=12&t=82833&hilit=Synthesizer
Rich, this is the original thread by Macrohenry. There is a follow up somewhere, where I organized the production of a batch of circuit boards from a company called FarCircuits.


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Thu 01, 2018 7:18 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3458
Here's a link to the freq synth info on my webpage. Be sure to read the whole thing when ordering parts, including the parts list at the parts placement diagram so you get an appropriate PLL.

http://www.tompolk.com//radios/frequenc ... synth.html


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Fri 02, 2018 12:12 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Rich; Got the transistors you sent. Thanks! Pretty much as expected the 3904 from my stock causes more current than the one you sent me. Harmonics and modulation are about the same. Slight flattening of the envelope peaks just before 100% which I think is fine.

The 2n4401 when swapped directly with no changes to settings or components produces more current than the 3904s but R2 takes care of that. The ringing on the RF waveform is lower with this transistor and overall I slightly favor it.

The good news is that with the changes you made all three transistors are workable and the difference is not a show stopper.

Since we do not have the LED feeding the ECS to peak the antenna I have have been using a 1 ohm resistor and just using the voltage drop across it as my tuning indicator. Also useful to monitor transmitter current draw.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Fri 02, 2018 2:36 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 513
Location: Equinunk PA 18417
Thank you both for links! A PLL synthesizer was used by SSTrans and I've seen other versions out there too. Very smart idea. Mac, does your cmos version work equally well on 5 or 9 volts?

Rick, my most recent schematic includes the tuneup indicator LED, full legal power, 9 volt supply. That is another argument in favor of supply higher than 5 volts.

I'm still messing around with this distortion thing but first wanted variable frequency. Today I built an interface to use a Heath IG-102 rf generator in place of the ECS chip. Conversion of its millivolt sinewave to a square wave 0-3 (or greater) volts. An AD844 into a logic gate gave me something pretty good that switches Q1 well, but is not always symmetrical. Symmetry is probably important to some degree. Symmetry shouldn't be a worry with the PLL synth. The ECS chip has a spec'd range, something like 45-55 percent worst case (and is hopefully much better). My interface is variable depending on gen level, so first order of business is to make a symmetrical square from it.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Fri 02, 2018 6:36 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3458
richfair wrote:
Mac, does your cmos version work equally well on 5 or 9 volts?

As long as you pick 5V ICs, it will run fine at 5V. As mentioned above, appropriate ICs are listed in two places on the original page. I can't remember if all are CMOS, but that will be easy to figure out when ordering parts.

The main thing is to avoid using the 74H series PLL. Seems because it's designed for higher frequencies, it uses different frequency setting and filter component values than the other ICs listed, which are interchangeable.


Last edited by Macrohenry on Mar Fri 02, 2018 5:03 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Fri 02, 2018 3:23 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 513
Location: Equinunk PA 18417
Yeah, you're right, I asked a lazy question. Looked through some datasheets this morning and got an answer.

edit: for the record, Macrohenry quietly added the bottom half to his reply AFTER my response

_________________
-Richard Fairbanks


Last edited by richfair on Mar Mon 05, 2018 4:37 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Sun 04, 2018 12:23 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Good news. Colin sent me the Macrohenry synthesizer that he built. Pretty much plug and play. I started by seeing how low I could go with the voltage. At about the 5 volt mark the synthesizer is not stable. So I started at 9 volts and just plugged it into the circuit in place of the ECS. It worked. I measured the output of the synthesizer with my scope and it was around 4 volts so I started increasing the supply voltage. I took the output up to 5 volts pp and that had the supply at 12 volts. Right where is was designed to work. Had to leave the 386 on the LM7805. It did not like the 12 volts. Need to go back and see what is going on there but I am running out of time.

Just wanted to give a shout to Colin for shipping the part to me. Thanks! Also Macrohenry; looks like we have a winner!

I will folllow up with some more tomorrow some time.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Sun 04, 2018 12:59 am 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1368
Location: Mount Pleasant UT 84647 (from Coventry, UK)
Rick, I am happy to help, and glad to hear that it seems to work. I forgot to mention that the supply was 12 Volts. I used it in a tube transmitter, and ran it from the low voltage winding of the transformer, with a bridge rectifier and 12 volt Zener.


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Sun 04, 2018 6:02 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3458
black85vette wrote:
So I started at 9 volts and just plugged it into the circuit in place of the ECS. It worked. I measured the output of the synthesizer with my scope and it was around 4 volts so I started increasing the supply voltage. I took the output up to 5 volts pp and that had the supply at 12 volts.


Glad it's working for you! QUESTION: Did you measure the output voltage with load attached or was that no load?

I have a request, too. Would you read the markings off each chip and post them in your response? Especially the 40103. THANKS.


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Sun 04, 2018 2:43 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Macrohenry wrote:
QUESTION: Did you measure the output voltage with load attached or was that no load?

I have a request, too. Would you read the markings off each chip and post them in your response? Especially the 40103. THANKS.


I did the measurement with load attached while in operation. Watched it on the Oscope to measure and to keep an eye on the waveform.

Parts on the board are: CD4060BE from TI, CD40103BE from TI, MC14046BCP from Motorola.

Currently have it running at 1500 KC.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Sun 04, 2018 8:20 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3458
black85vette wrote:
I did the measurement with load attached while in operation. Watched it on the Oscope to measure and to keep an eye on the waveform.

Ahh, that makes sense. I was thinking the output should be very close to the supply rail because the counter/divider should output either zero or one. The circuits I've been working with put a high impedance load on it, so my measured voltages are higher.


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Mon 05, 2018 1:22 am 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Had some more time and found that both the synthesizer and the 386 worked well together with both at 9 volts. Eliminating the LM7805 from the circuit. Even though the synthesizer will work, the output of the transmitter does not work below 7 volts.

The final output sine wave distorts slightly with antenna tuning where the original mostly just changed amplitude. There is a point at which it is matched and the waveform is good. But tuning this transmitter is not as simple as the original design and just peaking the LED. I was looking for peak plus best looking wave form. The tuning capacitor had a major effect on this. Even a small amount of capacitance had an impact on the waveform. Tuning with a variable inductor allows peaking the signal with minimum capacitance added. With this being frequency agile I don't think a fixed inductor will work unless you just leave the frequency set after you optimize the antenna tuning. I am not certain but the synthesizer may be better off with a buffer stage to isolate it from the transmitter. Any suggestions for a high impedance input with an output similar to the ECS?

I got this all running and then added in the audio compressor Macrohenry designed. Now running at 1500 KC with audio processing up front and the ability to dial the output power up to full or down for nearby receivers. Works for me and well enough to order the synthesizer parts to build my own. The synthesizer is awesome for our needs since it tunes the AM band in 10 KC increments and is really stable.

If anyone has a measurement/adjustment to be made or suggestion on this let me know so I can test it ASAP and get the synthesizer back to Colin.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Mon 05, 2018 2:34 am 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1368
Location: Mount Pleasant UT 84647 (from Coventry, UK)
Rick,

There is absolutely no hurry on that synthesizer. Keep it as long as you need it in order to perfect the system. Ultimately it helps us all out.

Colin


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Mon 05, 2018 4:54 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3458
black85vette wrote:
I am not certain but the synthesizer may be better off with a buffer stage to isolate it from the transmitter. Any suggestions for a high impedance input with an output similar to the ECS?

I began to wonder the same thing when the P-P output under load in your circuit was noticeably less than the supply voltage. I'm feeding a FET, so it measures the same with both load and no load.

A buffer could be as simple as a transistor with a couple of biasing resistors. It may not have the same voltage output, though. I'd have to think about that now that I think about it.

If you don't want to experiment, since you're already using a dual supply for the audio processor, you can simply add a fast op-amp follower such as the LM318. The NE5532 is speced up to 10MHz but it doesn't perform well as a follower in this case.

Colin and Richard, thank you for being so generous with your parts. It doesn't go unnoticed that your are helping produce some good work.


Top
 Profile  
 
 Post subject: Re: Variation on the 35Z5 ECS oscillator transmitter
PostPosted: Mar Mon 05, 2018 4:33 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 513
Location: Equinunk PA 18417
Thank you Mac.

Rick, I was PM'ing you when we lost power a couple of days ago. Everything is back to normal now.

A couple of things come to mind. Q1 benefits from a square wave on its base that switches fast and goes near zero. It seems the actual voltage of logic high is less important, 3 volts or higher seems adequate. (Q1 will draw current through R1 when the square is "high", so the value of R1 cannot be ignored.) The ECS oscillator is a 5 volt part that does those things. The MC14046B PLL is cmos and, like ECS, is spec'd to source logic low at 0.05 volts above ground, which is low and good. So, you'd think Q1 would be happy with the cmos output and yet yours is not. Is the synthesizer output as clean as the ECS output? CMOS supply can be between 3 to 18 volts, with switching speed dependent on voltage. That PLL's oscillator is guaranteed only to 1mHz with 10 volts supply, and with only 5 volts just half that (500 kHz). Not sure if that is affecting you.

Just before our power loss I had built a buffer for my Heath rf generator so I can transmit on difference frequencies. After applying a bunch of gain I have a somewhat dirty square wave from +1 to +4 volts. Q1 worked with that signal, sort of, but poorly at best. I shaped that through a standard TTL Schottky gate for much sharper transitions with near-zero lows, and Q1 was happier. I will make more progress today, finally.

_________________
-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: Dare4444 and 3 guests



Search for:
Jump to:  
























Privacy Policy :: Powered by phpBB