วีดีโอ

Hi Brendan, thanks for the information. Very good on the null - excellent! Regards, Lou

Hello, thank you. Cheers.

Ok. Thank you. Cheers

Hello, the amplifier came with some Teflon covered wire about 1.0mm diameter. I did not use that because it was too thin. I used some multistrand Teflon covered wire about 2mm diameter of wire, on one amplifier, and the one in the video was 2 mm enameled copper wire with blue heat shrink for better heat-resistant insulation. Just to see if heat shrink is as good as Teflon. Yes, it is just as good because it was 450 degrees rating and easy to get compared to Teflon. The output RF transformer primary consists of 2 copper tubes inside the Ferrite cores. They go to the drains on the left and are joined on the right, bypassed and connect to +12VDC. It is half a turn only. The tubes came with the kit. The secondary is 3 turns of the wire described above. At 50 watts RMS, the peak-to-peak voltage across my 50-ohm dummy load was 50V rms which is 70.7 V peak, and 141.4 volts peak to peak. That is all I got from that amplifier at 7MHz with -30dB intermod distortion. I did measure a maximum of 78W at 7MHz, but the distortion was bad. And I got 30W at 28MHz, with -30dB distortion. The input RF transformer is 2 turns primary using some Teflon covered wire from a piece of thin coax, and the secondary driving the gates is half a turn using small copper tubes inside a smaller core. Same idea as the output transformer but smaller core. These details came on the circuit I got from the kit supplier. You can find that circuit on the web called the "MiniPA70" amplifier from China.

Hi Austin, very good. Thank you. Cheers.

Thank you for the comment. Yes, the video was made to illustrate the different types of small connectors that can be used on homebrew equipment. The first section was requested by students as part of the decision-making process and how to distinguish the different pins. I am glad you found the later section on how to make up leads without a crimping tool, and the removal, helpful. Feedback is always welcome so thank you for taking the time to express your view, Enrico.

Hi, thank you for the thumbs up. There is another video which I made after adding a PE4302 programmable attenuator. I use mine a lot for testing RF circuits and sweeping over filters both slowly and fast - cuts a lot of design time down. Cheers here is the link for the PE4302 demo. th-cam.com/video/XAnesvP3078/w-d-xo.html

There is more here on the generator. Give me your email and I can send the code and circuit. Cheers, th-cam.com/video/XAnesvP3078/w-d-xo.htmlsi=LjUOqvc3urpKYkou

Hi, thank you. Yes, I made it so I don;'t have to climb the tower. Not high but good enough for what I do on the air. I am on a moderately high piece of high ground which helps. Cheers.

@@vk3aqz absolutely you made life easier for sure and it will get out awesome on air too good on you mate thanks for replying 73s have. Great day .bill

Hello,thanks for the compliment. The Hantek I have is a 200MHz one and has an FFT function. I imagine yours does too as you mention. The FFT function is selected using the "Math menu" button between the CH1 and CH2 select buttons. That should bring up the "purple" FFT frequency spectrum display. It has a menu which allows you select the channel, window processor (I use Hanning - reduces leakage better around each FFT sample), dBm or linear scale, and FFT zoom EG. X1, X2 etc. I use X5 which is good for spreading out the 2 frequencies corresponding to the 2 tones you are feeding into the transmitter audio input. The Probe will be on the RF stage you are testing and you will have set the CH1 gain so that you get a good size RF spectrum at the carrier frequency. You need to use dBm for the scale. Then you adjust the horizontal position, and the timebase rate, turning the horizontal position slowly, till you see the 2 vertical lines of the tones appear normally from the left. They should appear down near the start of the spectrum and are essentially a product of the scanning frequency and the FFT samples. You need to adjust the zoom to move, or spread the 2 tone spectrum lines. They will appear down near the start of the FFT and by adjusting the horizontal shift you can bring the tones into view. I do have an analogue spectrum analyser (HP) which also can display the 2 tones and I find the Hantek is pretty close. However the dynamic range of the Hantek is around 60dB whilst my HP is around 90dB. But for ham radio testing you are looking at around -30dB, or better, IMD in the main signal. If you are trying to test the receiver IP3, say at 14Mz between two RF signals 10KHz apart, then you need a totally different setup (see the book Experimental methods in RF design by Hayward, Campbell, Larkin etc for more info). Cheers, Lou

​@@vk3aqz thanks for the fantastic reply! Up until now I have not been able to get the same level of resolution or refresh rate speed as you, from my FFT mode. I will try your suggestions and will report back. One more question - in the FFT screen I see at 3:15 in your video, what is the center freq and span? Is the center freq the RF carrier freq? (e.g 14.100Mhz) or is it the audio freq (e.g 1 to 3KHz or there abouts? ) I'm trying to get my head around TX IMD testing, slowly but surely :) . 73

@@mostlypostie1It is some time since I used the FFT function. I think the when you select FFT, the knobs do different things to the normal oscilloscope function. The timebase knob changes to , I think, samples per second. And it changes the spectrum width as well. The 50Hz at the bottom is probably the sample rate (Hz per second I think). The timebase window at the top might be the spectrum width per division. I think it is 800us, which is around 1.2kHz per division. The 2 tones I was using were around 650Hz and probably around 1.9kHz which means they would be around 800uS apart on the display or 1cm. The actual RF frequency was 3.6MHz. Something that happens is the you can get aliases and foldback when there is RF above the Nyquist frequency. The foldback is caused by undersampling so you might see the 2 tones at several spots inside the spectrum width. The X1,X2, X5 and X10 is a magnification function of the spectrum (you only see more or less inside the screen width). So the spectrum spreads out or gets narrower hence the term FFT zoom. The main thing to think about is that the normal CRO controls do something different and if you look at the knob markings, you might see they contain 2 types of functions printed. It's not all that clear but I think the manuals have more info when using FFT - in FFT mode it is no longer a CRO really but a spectrum analyser and the knobs become spectrum controls instead. Maybe feed a 2 tone RF into the CRO . Say 3.5MHz, and play with the knobs. Make sure the 2 tones are not harmonically related or you won't see the IM products since they will be on top of each other. See how you go. Cheers.

Yes, input was only a few hundred mV pp and gain low so only from 1 to 2 volts pp out. Rails 12V. Only the poor quality LM358 show the distortion. But I also did all sorts of tests at other levels to. It made no difference to the distortion. And good brand LM358 ICs are clean. It might be caused by phase reversal due to intrinsic diode like junctions in the die as described in engineering literature from Texas Instruments. Poor quality materials can apparently contain intrinsic semiconductors such as diodes and transistors inside the substrate. As this batch came from an Asia source , the quality and cleanliness of the materials is questionable. Looking closely at the waveform down in the negative part of the sine wave, we can clearly see the voltage do a small change in direction which is a phase change. This effect tends to show up during the low signal current part of the signal. Looking at the waveform you can also see the onset of the conduction points in the intrinsic semiconductor on each side of the phase reversal portion.

Hello, yes, I have used these OP amps for many years. In this video I am demonstrating the distortion produced by a poor quality one purchased via a local supplier who buys from Asia. The tests I have done, include some circuit changes such as varying the source resistance (the source resistance has a significant effects on the THD ), and load values. In this demo the input level is quite low and raising moderate gain. The output level is around 1 to 2 volts pp and does not go anywhere near the rails which are 12 volts. So we are not clipping at the rail. Varying the input level makes no difference to the distortion. The low quality LM358 shows a waveform distortion at the most negative point of the waveform. It looks like one of the junctions inside the package is not behaving correctly. These devices are balanced amplifiers with constant current sources. I suspect the device meant to produce current balance between the long tailed pairs is a poor quality material. I know that LM358 chips can achieve at least -60dB THD (0.1%THD), and I also use FET chips like the TL072 which achieve figures down to 0.003% THD. In this project I only need -60 dB. So this is just a comparison between a good op amp, and a bad LM358. When I did the video I did have an LM833 in one channel but it was the last test I did so I left it in. Earlier tests with good LM358s (NEC and other brands) showed -60 dB or better with the good LM358 and only -40dB with the poor quality one. I have a tube of 30 bad ones and all the ones I tested from that batch were bad - clearly a manufacturing issue not a design problem with the LM358.

There is a good "Application Design Guidelines for LM324/LM358 Devices" paper written by Ronald Michallick for Texas Instruments in 2019 and updated 2021. You may have already read that paper. It describes in detail the design of the chip including the various currents. Although the original design was many years ago, there are still modern versions manufactured. The SMD version is just one with improved performance. These are common in quite a lot of current equipment. At the university I was tutoring at, we had manufacturing dies for many ics. One student Engineering project was to examine the LM358 design and then produce a working chip in our ovens and machinery. I believe the effect I am seeing in the poor quality LM358 from Asia may be "Phase Reversal" with a negative going signal. The input signal current activates parasitic NPN transistors which then takes current from other internal connections and inverts the signal phase. I believe that is what I am seeing in the trace where the signal appears to turn upwards then downwards and then up. Poor quality semiconductor material can contain unintentional diode or transistor type junctions.

Hi, No webpage is no longer working. Was costing big money to the site managers each year ($800) so I just use Facebook and TH-cam now.

Thanks for the compliment. Yes, my XYL is also VK33ZY, Rhonda. She was licensed around 1968. The tower works really well. I designed it a few years back,. The height meets the council tower limits here which only let you go as high as 9 metres or 3 meters above roof. The A frame tower you see is 4 meters but can also be made another 2 meters higher. The tower calculations will allow the centre pole, made of scaffold grade Aluminium, 3mm thick, 65 mm diameter tubing, to go 4 metres above the tower top. The strength calculations allow for a full size 40 meter, 3 element beam. The tower itself, however must be guyed, or supported, by a wall, as seen in my video to meet those ratings.

Yes, the IC706 I have was the first one in Australia. I worked at a ham outlet at the time and it was used in a magazine review. However I build a lot of my own gear so I don't buy much commercial stuff. I have been a ham for 55 years and built gear all the way up to 14 GHz (For S band experiments to the geostationary satellites used here in Australia for the outback and remote islands television services when I was the Chief engineer in a TV network).

Hi Jack. It measure 12.6dB and beamwidth is around 30 degrees. It is mounted at 6 metres above ground. Regularly works upto Alexandra in northern Victoria around 250 Km away using 30 watts SSB on the 2 metre band (5 by 5 to 5 by 7 signals both ways). And Northern Tasmania at times around 400 Km away over Bass Straight.

Hello Alan, thanks for the compliment. Yes, unfortunately, the PC screen interferes with the camera focus system and sometimes I don't notice that it has gone out of range. The actual camera file is high definition but it is way to big to post on TH-cam so I have to down convert to MP4, which adds additional softening. This software is part of my MK5 transceiver which is currently waiting to be published in the WIA AR radio magazine. I used to put some software on GitHub but I found that people started stuffing around with the original code till it ended up full of bugs but users were contacting me to fix it. Some of the code that was added, consisted of one line commands with lots and lots of commands in the same line making it impossible to follow! And GitHub also wanted me to pay to put the stuff on it as well, and were becoming a pest. So now I only send the code directly to people who know how to use it when it is finalised, and through my website(which is currently been rebuilt). If you like I can email you the ino of the TFT meter file. Send a message to destefano@dodo.com.au Regards, Lou

@@vk3aqz Thanks Lou I have dropped you a email - regards Alan

Hi Thank you. Cheers, Lou

Hi Magikflame. No, there seems to be some issue with notifications on many posts now. And our internet here is now so unreliable that we only get an hour or 2 a few times a week due to storms and power outages.

Hi Brendan, yes that version of Jim's SWR bridge is pretty good. Simple and reliable. Ok on everything else. Take care, cheers, Lou

Hi Thank you. Cheers.

Yes, that is around 35% in the container shown. Thank you for your comment.

Hi, mine measured around 135uH. so its reactance at 3.5 MHz, the lowest band my linear goes, is nearly 3000 ohms. If your linear goes down to 1.8 MHz, you need around 220uH. The reactance has to be high enough to make sure not much of the RF goes down through it, and lost at the bypass caps at the bottom. It really needs to be at least 3 to 4 times the impedance the PI coupler puts across the tube after converting the 50 ohm cable load back across the tubes. The RF plate impedance of the tubes must also be lower than the reactance of the choke but that is hard to calculate. It's not the DC voltage by DC anode current, but the RF voltage by the RF current, which depends on how well loaded the tubes are into the Pi Coupler, and any phase shift between the RF voltage and current. And in SSB mode it goes up and down with speech peaks. The 135uh works for me okay. Make sense?

@@vk3aqz Hi thanks for your reply. The amp covers 80m 40m 20m 15m 10m and I will try for 135uH I have a antenna analyser also doubles up as a GDO which will give me the inductance or could use my inductance/ transistor tester/ capacitance meter .

@@itsonlyme9938 Ok. Yes, the analyser should show where the resonance frequency is. I checked it with the 2 ends of the coil connected together mounted in the case (as per my video). The enclosure metal will change the resonant frequency so most people advice on doing it in the box. I actually did a measurement outside the box on its own just to see what it was. In the box, I think it changed by around 100kHz but I cannot remember if it went up or down. So maybe do the same first. The wire size you are using should be fine. I think the current could be around 600 mA. I actually took about 6 feet or thereabouts (really needs to be about the same length as the choke) of the wire and passed an amp through it using a 12 volt variable supply with a 10 ohm resistor in series (10 volts across 10 ohms = 1 amp). The actual resistance of the choke wire is nearly zero so you need the resistor. About 20 seconds is long enough. If it blows it's too thin -- Hi !! Use thicker, or 2 strands in parallel if you don't have thicker wire.

Hi Graeme, thank you. Cheers mate.

Yes - simple tools do work!

Hi Graeme, thanks for the compliments. Cheers.

Hello, the design has been sent to the Wireless Institute of Australia for publication in the National magazine so parts of it now belongs to them. The details will then be available in the magazine and the software too. Should be a few months away. I will then post here and contact you. You can visit my website also for more videos and contact details. www.vk3aqzkits.com Cheers and thank you for your enquiry.

Hi. Thank you. Some of the recent changes in Electronics, such as the small processors, and SDR chips, now been used to make transceivers and other ham gear, are stimulating new interest in ham radio. So I have seen quite a few of the "older" guys get back into radio. TH-cam is also helping. Cheers.

Hi Jerome. A multiturn trim pot would be heaps better. I recently built a 10W amp using RD16HHF1 fets and used a multiturn pot and it was a lot easier. Just one thing to be very careful of when adjusting the bias trimpot. Do not go all the way up passed about 4 volts, or so, or you will damage the FET gates. Keep an eye on the drain current. A 2 tone test oscillator, and some means of checking the intermod distortion around the carrier, is a good idea. Otherwise adjust the bias so that you audio does not get raspy which will happen once you get intermod levels lower than around -25dB. As far as the harmonics go, that amplifier is using switching FETs, which basically produces a sort of square wave rather than a nice pure sine wave. So you will always get high harmonics out of those amps. You must have a low pass filter on the output of the amp. Something that gives at least 50 dB down on the 3rd harmonic. The even harmonics will not be as high since the amplifier is a push pull design, and that tends to produce lower even harmonics. I use 7 pole filters on all my solid state amps, including the transistor ones, such as the MRF454, and the RD100 FETS. One thing to note is that the RD series of FETS are designed to be linear at RF, whilst the IRF types have a transfer curve designed to switch fast and not as linear. So the RF FETS give lower harmonics, and are designed to handle much higher SWR levels without blowing up. The cheap IRF FETS are a lot more delicate and you really need to make sure the SWR is low by use of some sort of protection, and by use of well designed 50 ohm filters on the output.

Hi. Yes thank you. Eventually most hams will need to make sure the RF levels from the transmitters do not cause injury of some kind, Ok on your test gear - sounds good. Cheers, Lou

Well - thank you for the very high and glowing compliments William. One gets pleasure from making a kit that people like. Much Appreciated. Best Regards and Take care, Lou.

@@vk3aqz..every syllable from the heart and justly deserved, mate!

I don't have a Palomar bridge so I don't know what the dial values mean. If you have the manual for it, it should be in there. Some bridges have a chart which tells you what value of the impedance is for a given set of dial values. The value of the R and the value of the xc together go to make up the value of the impedance - Z ohms. On my bridges, the dial values get put into the computer program I supply with the bridges, and that gives the actual impedance value as R - jXC or R + jXL. I can only guess that the Palomar R value is the REAL component of the impedance, and the Xc value is the amount of capacitance in pf - that in turn needs to be converted to capacitive reactance at 3.6MHz using the equation Xc = 1/(2*pi*f* C) ohms. That in turn would mean an impedance of R - jXc, which would be 25- j 4420.9 ohms. The R value of 25 ohms makes sense but an Xc reading of 10 does not sound right - but it could be!!. So you really need to read your manual to find out what your dials are actually representing - particularly what an xc reading of 10 represents. Cheers.

Not sure why I bothered to answer this question - no feedback!

Thank you very much you answered my question I found the manual and read it 73s my good friend

@@Kc8qdq Ok. Very good. Thanks for the reply. Cheers jeramy.

Yes, the diodes were all replaced with 3Kv 6A ones after the video. The original square ones are still good but new ones are better with a lower voltage drop so run cooler.

@@vk3aqz How many 6A diodes did you put back in?

@@michaelmacleod 2 on a bracket. They are stud ones from a large solar 5kw inverter. I think they are made by Semikron.

I should add that after I removed the original diodes on the PCB, I did replace them with 1N5408 diodes and new equalising ceramic caps. Just in case the heavy duty ones resulted in too much inrush current into the new caps and over stressed the HV secondary wire. So far the heavy duty ones are working but I think I will put back the string of diodes which have a slower turn on time, and reduce the inrush current. So in effect I am not all that comfortable with the heavy duty inverter diodes just in case the transformer secondary gets overstressed. And I believe putting a fuse in the HV line to the diodes is apparently not a good idea either - can cause flash over in the tubes if it blows, I have been advised!

Thank you.

Hi , On the video I am using the FFT section of the Hantek DSO5202P simply for illustration. I also have some older RF spectrum analysers including a HP8568A but it is too big and bulky for the video demos. For intermod tests the 2 tone signal is from a VK5JST low distortion oscillator published in the AR magazine. You may know the following but I will just note it for any readers of this post. The 2 tone oscillator should have the harmonics of the tones around 50dB down. The spectrum analyser should be able to also measure the noise around the intermod products signal as an RMS value, and the analyser needs to remove a factor from the overall result. Not many can do that. The dynamic range of the analyser needs to be better than 80db. The analogue types are capable of that type of performance. Digital ones need to really have something like a 12Bit, or more, A to D. The Hantek only uses an 8 bit A/D so it cannot really be better than say around -40 dB. I use the HP for more accurate work. A good second hand analogue one is better than the low cost digital ones. If you can find an RF spectrum analyser that can measure down to 10Hz bandwidths, that will good. As mentioned above, the analyser should really be capable of also measuring any noise inside the bandwidth of each intermod tone and factor that in. The noise may be due to a noisy transmitter PA or other parts, or maybe a noisy log amp in the analyser. In the case of a digital one, the noise may be quantising noise which will give a false value and effectively make the result worse which cannot be blamed on the transmitter.

See reply to your earlier request. Please note the internet to me is very unreliable dropping out for days on end. So I can only reply when it drifts in and out of range.

Thanks for returning...i am only a beginner at aligning ssb cb radios..i worked out that your 2 tone signals were emitting from your tranceiver..i have a few analogue Sig generators Marconi2019, Boonton102A, and Logimetrics 921a which have much better variable output but as you said take up way too much space so i bought a FY6600 , which require a bit of attenuation on its 1mv output..just trying to work out how to send the 2 waveforms from ch 1 and ch 2 from the FY6600 to become tones then through the front end of the cb,,...as i said i am just learning.ps so undecided which spectrum analyser to get considering in only need it for the cb band.. Much appreciated.

​@@andrewverran3498 Hi Andrew, yes all oaky. We were all beginners once! The 2 tones can come from your tone oscillators and must be of a frequency that both can pass through the SSB filter width of your CB radio - typically 3Khz or so. The tones must not be harmonically related. You might use 700Hz and 1.8KHz which are 2 common tones used. The 2 tones are combined into one signal using a small audio mixer or just a simple potentiometer. One side of the potentiometer is connected to one tone active lead output, and the second tone to the other side. The middle, or moving arm, goes to the microphone input of the CB where the hot or active wire from your microphone would go. The earthy ends of the tone leads join together and go to the earthy connector on the CB radio microphone input. NOTE: if you have a powered microphone, there will be a DC voltage on the mic input pin (Bias). So use a capacitor (say 1uf) between the tones and the mic input or you might damage your radio! The level of the tones should be at the same level as the peak of your speech coming out of the microphone. The distortion is worse at the higher levels so the test has to be done at the peak level of your voice. The potentiometer moving arm is used to adjust the level from the 2 tones so that both tones entering the microphone socket are the same. Moving the pot arm back and forth will increase one level or the other. It is the tone balance adjustment - very simple method - I use that method. So feeding the combined tone into the microphone socket you then wind you mic gain up and down till to see the average power on a meter at about half the maximum say 7 Watts for the SSB setting. That's because you have essentially 2 carriers in the pass band transmitting, each of equal power - so one tone at maximum level might give you 14W (typical SSB CB radio output) whilst 2 tones should show half power with the 2 levels at maximum. The waveform has 2 mixed sine wave patterns instead of one - you might have seen a 2 tone test waveform on CRO. And with speech, an average power meter might only kick up to maybe 3 or 4 watt ( but the peaks will reach 14W - which you won't see unless you have a peak reading power meter!). I hope that helps. If not maybe ask someone near you who can help. Cheers.

@@vk3aqz thank you so much...great advise...i have printed it out for my records.

Hi, yes, thank you for all that. Currently in the desert travelling. Hope to be back in civilisation soon. Cheers and stay healthy.!

The noise bridge can be used in many ways. One can use it to preset the tuner so that an antenna that cannot be adjusted for any reason, will look like 50 ohms to the transmitter. It also means you don't need to transmit in order to adjust a tuner - some commercial tuners actually use an internal bridge for that very reason. A tuner and bridge is also used by SWLs to get the best out of a random long wire - much easier for an SWL trying to adjust a tuner by trying to listen to atmospheric noise only. And yes, you can also use the bridge to trim an antenna so that it becomes 50 ohms and no tuner is needed - but that's just a basic common way to use a bridge, which is a well known usage. As I said, the noise bridge is really an RF "ohm-meter" (i.e. to measure an impedance such as z+jx or z-jx) or "RF multimeter" used for all sorts of RF measurements - and even as a vector network analyser using good old basic radio theory maths! (Digital VNA's use the same circuit - only it uses a microprocessor for the maths and sweeping the circuit!).