The AM Forum
March 18, 2024, 11:51:55 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: 1 ... 6 [7]   Go Down
  Print  
Author Topic: Building my first oscillator  (Read 151732 times)
0 Members and 1 Guest are viewing this topic.
W4AMV
Member

Offline Offline

Posts: 672


« Reply #150 on: September 20, 2011, 02:17:26 PM »

Hi. They may be MOLEX, but the exact part no. etc... not sure. You might email his address and ask.
Logged
ssbothwell SWL
Guest
« Reply #151 on: September 20, 2011, 07:21:43 PM »

i just shipped my mixer. hopefully it will arrive relatively soon. hopefully the w8diz mixer will arrive relatively soon as well.

what do you think i should work on the in the mean time? i need to put the LO oscillator together on copper but i want to make sure i have the right dBm output and the right frequency range. should i aim for 0dBm?

if in the future i wanted to make oscillators that didn't require large coils what would be my options? crystal oscillators, pll synthesizers? anything else?
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #152 on: September 20, 2011, 08:19:53 PM »

ok, fine I will look for it, set it up, and report back. On the LO, goal should be 7 dBm nominal and good stability and of course reasonable spectral purity. Not sure what your desired receive band is and goals. I would keep it simple. Say the broadcast band! However, if you want to go shortwave, pick a favorite frequency range. Your initial 160,80,75 meters seemed reasonable. If you want to start investigating a  PLL, synthesizer, etc... I suggest you start a new thread! I am sure there are experts on the forum who will provide a lot of thoughts in this area.

Alan
Logged
ssbothwell SWL
Guest
« Reply #153 on: September 20, 2011, 10:11:39 PM »

maybe i should stick to one project at a time.  Smiley a pll synthesizer can wait.

i would like to make a broadcast band receiever but i have noticed that the lower i drop the frequency range, the less total tuning range i get. with it operating around 1-2mhz i only get about 0.5Mhz of range. perhaps i dont have a high enough inductance coil but i haven't even gotten it to oscillate under 1.3mhz.
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #154 on: September 21, 2011, 01:49:34 AM »

yes. Here is a good bedtime read.

http://dspace.mit.edu/handle/1721.1/15991
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #155 on: September 22, 2011, 09:13:18 PM »

Hi, I received your unit in good shape. I checked your mixer and it does function, however, there is a problem. Your loss is high and the balance is poor. This is the reason you are having difficulty in observing the IF. So, where is the problem? I checked your diodes, they are fine. I believe the issue is your transformer connections. Since you did not use colored wire, a trifilar xmfr is easy to get wrong. So, I am going to start there. 
Logged
ssbothwell SWL
Guest
« Reply #156 on: September 23, 2011, 02:50:05 AM »

weird. i'm pretty sure i put the transformers together correctly but maybe i mixed something up in the secondarys (bifilar side).

thanks for checking it out.
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #157 on: September 23, 2011, 09:49:02 AM »

Most likely or used one of the windings meant to be a secondary as the primary. The balance and feedthrough of the LO and the RF to the IF is pretty significant. In a reasonably balanced mixer, these are doubly balanced, the LO and the RF level are reduced significantly with respect to the IF. So, shuting down either the LO or the RF, will force the IF signal to goto nearly zero. There would be no output whatsoever. Not in the case of yours, the LO is quite well coupled to the IF port. The diodes are spot on. The Vd nearly 0.3 V at 2.5 mA and they are well matched.   
Logged
ssbothwell SWL
Guest
« Reply #158 on: September 23, 2011, 05:25:31 PM »

the way i checked the transformers was to do a continuity check on the ends. the two wire ends that have continuity with each other but not the other 3 wires are the primary side and center point of the secondary side is where there are two wires soldered together.

are the two non-center tap wires of the secondary coil identical electrically or do they need to be distinguished from one another in the circuit?

edit:
my W8DIZ diode ring mixer kit arrived today. i just assembled it. i cant test it until i get a new fet for my LO oscillator. hopefully that will arrive tomorrow.

i also got an IF amp stage kit which is assembled. these W8DIZ kits are pretty nice. if only they had an VFO kit, then i would be able to build an entire receiver from kits. 
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #159 on: September 25, 2011, 10:04:19 AM »

Hi. I will try to check the xmfrs today and let you know. It may be latter this week as I have a couple of other items. 
Logged
ssbothwell SWL
Guest
« Reply #160 on: September 25, 2011, 02:15:24 PM »

okay. take your time and let me know how it goes. i got a cheap frequency counter at the swapmeet (dsi 5600A). I'm waiting for the 9310 fets to arrive in mail before i can do any more work with the LO or test my W8DIZ mixer.
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #161 on: September 25, 2011, 03:15:49 PM »

Hi. I looked at the xmfr on the LO/IF side first. The SINGLE and ONLY center tap as you recall is the IF output. The start-finish winding of one set of wires is the LO primary. The other TWO sets of remaining wires, properly connected create the CENTER TAP IF AND the two remaining wires drive one half of the diode ring. For some reason, you connected ON THE SAME SIDE of the start-finish winding the center tap for the IF. In other words, you did not connect the START FINISH of wire pair but in fact the START-START of a set of wire pair. The main bad result of this is incorrect phasing of the LO voltage to the diode ring and thus poor balance of the LO signal to the IF port. This is an easy error to make. By START FINISH I mean observing the wire entrance to the toroid (the START) and its exit on the OTHER SIDE of the toroid (the FINISH). You had them connected on the SAME SIDE...oops. So, I will make a better drawing than what you used and post. Make the changes to the units and see the resulting improvement.
Logged
ssbothwell SWL
Guest
« Reply #162 on: September 25, 2011, 05:08:27 PM »

i was worried about that. my question "are the two non-center tap wires of the secondary coil identical electrically or do they need to be distinguished from one another in the circuit?" was my attempt to ask about the start/finish ends of the coils but i did not have the right terminology.

at least i now know what i did wrong. Smiley


when i made the transformers, i didn't realize there was a difference between the start/finish ends of the coils. i guess i need to be way more careful with my windings and keep better track of each wire and its start/finish ends.
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #163 on: September 25, 2011, 08:56:02 PM »

YES, indeed there is a difference in the START-FINISH ends of the coil and the resulting transformer. SO, I will prepare a drawing and I hope an understandable explanation why this is so.
Logged
ssbothwell SWL
Guest
« Reply #164 on: September 26, 2011, 03:37:35 AM »

i was looking at various oscillator schematics in the arrl handbook and 'experimental methods in rf design' and noticed that a lot of designs use toroids instead of large air coils. i replaced the large ceramic coil form in my LO with a T94-6 and it seems to perform perfectly fine.

what is the reason to use air coils instead toroids in oscillator tank circuits?
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #165 on: September 26, 2011, 12:25:04 PM »

A good question and you probably should post this as a new thread item to gather more views.

A quick answer, the TORUS and the SOLENOID are just two different physical geometries. So on the surface, there is a difference only due to their geometry. The torus is self shielding, therefore, the influence of the enclosure on inductance change would be less. The solenoid can achieve a very large Q and thats a benefit in oscillators for various reasons. However, the toroid can do quite well in achieving high Q, 200-400 is not uncommon. Toroid permeability must be carefully selected -6 material etc... check their temp. coefficient. Air inductors (solenoid) can achieve a very low TC (but it still has a TC) and maybe easier to temp compensate.

Again, there is more to this, but ping the general audience for more replies via a new thread.
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #166 on: September 26, 2011, 11:08:17 PM »

Hi. Your mixer works perfect. Now, both xmfr's were incorrect.
Logged
ssbothwell SWL
Guest
« Reply #167 on: September 26, 2011, 11:38:45 PM »

oh wow. is it balanced well enough to operate without the low pass filter? is it comparable to the hp mixer of yours?

i got my bag of J310 transistors in the mail today. i get a lot more gain. using j310 in the oscillator and the buffer amp i get 5.1VRMS signal into a 50ohm dummyload from the buffer amp output! why am i getting such a high voltage output? using a j310 in the oscillator and an mpf102 in the buffer amp stage i get 2.44VRMS output from the buffer amp.

arent these values really high? should i really be getting 500mW output from a j310 buffer stage? the transistor itself is not hot or anything.

the output from the 2n3866 stage when using the j310 transistors is 7.3VRMS.


in other news i put together some basic test equipment. here is a photo of a 10db attenuator, a return loss bridge, and a 50ohm 4W dummy load. i used bnc connectors and these neat little 2x2 enclosures. i dont have enough bnc cables to test the return loss bridge just yet. http://i.imgur.com/Wl28M.jpg sorry my desk is covered in paint and junk which makes it a terrible backdrop for photos.

edit: oh i am really dumb. i totally messed up when i put together my dummy load. i spaced out when soldering it and made it 50ohms pad in series with the circuit. oops. i'm fixing it right now.

okay so i fixed my dummy load and checked my outputs again. now i am getting 4.1VRMS from the 2n3866 amp stage. this seems a lot more accurate.
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #168 on: September 27, 2011, 04:14:12 AM »

Yes, the balance is quite good and no LPF is needed to see the IF. Both LO and RF signals are suitably suppressed. This is sort of a nice mark to qualify a mixer. I would need to measure balance value on a spectrum analyzer in order to bench mark it against the HP. The HP has additional xmfrs to further improve balance, so I would not be disappointed in a slightly worse number. The j310 is a nice device with decent gm.
Logged
ssbothwell SWL
Guest
« Reply #169 on: September 27, 2011, 07:23:45 PM »

very cool. i'm excited to get build a complete superhet receiver. for now i am going to use the w8diz IF amp i bought but i would like to try building one from scratch myself in the near future.
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #170 on: September 27, 2011, 09:43:03 PM »

RF-IF and LO-IF rejection (balance) is 50 dB, quite good. This is not surprising as the diodes are well matched. The xmfrs are proper now. If I have a chance, I will measure vs. frequency and that should do it.
Logged
ssbothwell SWL
Guest
« Reply #171 on: September 27, 2011, 10:06:09 PM »

i picked my diodes by using the diode test mode on my multimeter and picking diodes that gave the most matched value. is that a suitable way to match them or did i just get lucky?
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #172 on: September 28, 2011, 09:53:34 AM »

yes. I used a TEK curve tracer and the matching was very good from off state through 2.5 mA of forward current. Of course AC matching could be off. That would be another measurement.
Logged
ssbothwell SWL
Guest
« Reply #173 on: September 29, 2011, 12:02:55 AM »

in the test equipment section of the arrl handbook there is a schematic for a 3.7mhz oscillator which they recommend for receiver testing. here is an image of the schematic: http://i.imgur.com/LOU7L.png

it is supposed to put out a +4dbm signal into a dummy load.

i have all the components other than the 2n5486 FET. i used an MPF102 and put the circuit together on breadboard (i know its bad) and am only getting 20mVRMS output. the signal looks okay but its really low power. in the schematic they note that you can replace the 2n5486 with a mpf102 but to expected reduced output.

would you expect to get 20mVRMS output? i tried a J310 and i get 25-30mVRMS. i was hoping to get atleast 1mW output. could the breadboard and long component leads be producing major loss?



EDIT: NEVERMIND. i made the exact same mistake as in my diode mixer. i mixed up the start-end leads of the transformer T1. its working perfectly now. Smiley
Logged
Pages: 1 ... 6 [7]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.059 seconds with 18 queries.