Ramsey AM 25 troubleshooting

I'm going to guess your Ramsey is acting up because you don't have everything hooked up.

It is not a good idea to operate the transmitter without an output filter because the signal will be rich in harmonics. I suggest you build the unit per the instructions and get it working before applying any modifications.

Ramsey folks report that well over 90 percent of the problems with their kits are due to bad solder joints or parts not being properly placed. You might also double check the frequency setting. It is easy to confuse the switch positions as being open when they are closed. Also check your calculation for the frequency switches settings. I messed this up more than once.

I recall that the L5 slug on my unit was almost all the way out at the high end of the band (above 1600 kHz) so this may be normal. Check that the H1 jumper is properly installed for the frequency you want to use.

Adjusted according to the manual Q9 will run very hot so check the temperature. It needs to be heatsinked.

Almost all DVMs are not suitable for measuring RF voltages so if you are using a DVM this may explain why you get no reading across the dummy load R. Usually DVMs work on AC up to 10kHz with some working up to 100 kHz. Check your DVM specifications.

The AM-25 is a nice kit to experiment with and the stock unit works fairly well. Keep at it and you will get it to work.

Neil

i'm using an oscilloscope to check the rms voltage at the load. i've now fully assembled the filter circuit. i've got the H1 jumper set right.

i believe i tuned the frequency properly. its set for 1710khz and my receiver (barely) picks up the signal on 1710. if i had the dip switches reversed then it would be broadcasting on 840 but i don't hear any of my signal on 840.

Q9 gets slightly warm but not particularly hot when the transmitter is turned. Q3 is totally cold. there is zero volts on the gate and source and 5V on the drain of Q3. does that sound normal?

the only way i am able to pick up any signal is if i hold my receiver within an inch of my antenna.

here is some new measurements taken with my multimeter now that i have assembled the filter circuit:

Q9:
Base: 6.2V
Collector: 14.9V
Emitter: 5.673V

Q3:
Gate: 0V
Drain: 5.673V
Source: 0V

U1 Pin 1: 7.1V
TP1: 3V

when i check the output on my oscilloscope with a 50ohm load i still get no signal unless i am modulating with audio. when i do modulate it i get a 80-120mV rms noise with no identifiable carrier frequency.

OK on using a scope. Not all know the limitations of DVMs thus the mention of this.

All the voltages you reported look OK except for that of the gate of Q3. It appears that no signal is getting into the gate. Since you can hear something on the receiver might mean that the synthesizer is working and you are picking up radiation from this circuit.

Check the connections for L6 and L4. Sometimes, if not properly removed, the enamel insulation will be in the solder joint and prevent contact.

Use the scope to trace the signal back from Q3 gate toward Q6. If the synthesizer is running you should find the RF signal along this path. The signal at Q3 gate should be a RF pulse with several volts positive and about .7 volts negative.

You had previously asked about C16. This is used to adjust the carrier frequency if needed. It doesn't change it much so if you don't adjust this the tx. will still work.

Neil

Thanks neil! you fixed it! i traced it down to one solder point on L6 which visually looked like a perfect solder but the enamel was not actually burnt away. i got lucky in that there was a little nub of magnet wire sticking out of the solder point and i accidentally bumped my probe onto the tip of the wire while i was holding it against the solder.

i resoldered that joint and now i'm getting 2.5V RMS output. 🙂

Nice to hear you got it running. I think you will have a lot of fun experimenting with it. Let us know of your escapades.

Neil

thanks! i'm already seeing both why people have a problem with the am-25 and why people enjoy it.it seems to have some pretty bad problems out of the box, but it also seems like a transmitter that is very open to experimentation and modification.

i am having pretty bad audio distortion problems. i have to turn the audio input way down. the degree i have to turn down the input depenends on the audio i am inputing. this makes me think a compressor would help a lot.

what do you think of the modulator modifications shown here: http://teslapress.com/am25.html

the article recommends replacing the TIP310 (Q9) with a 2N3055. have you tried that?

I have noted that lowering the U1 pin 1 DC voltage improves the audio. This also decreases the final input power but it can be lowered quite a bit before the 100 mW limit is reached. Adjusted according to the Ramsey manual the final input power is greater than 100 mW. The change needed to lower this voltage for 100 mW operation is to short R29 permitting a lower voltage setting via R23.

The audio waveforms I posted at
http://www.part15.us/index.php?q=node/2714
were taken from the transmitter with the modifications in the post made to increase the efficiency. These were very good waveforms, the only problem being the flat lining at the low voltage valleys at 90% modulation. I think the diode added in the last link you posted is intended to extend the modulation to 100%. This limit is there because, contrary to the data sheet, the LM358 does not provide an output voltage down to 0 volts. I tried a diode in series with R33 and R34 (band toward Q3) to drop the voltage toward zero and the modulation went to 100% but I left this out since it also introduces non-linearity. Perhaps placed in the emitter circuit of Q9 where it would be in the audio feedback loop would work but I didn't try this. The main reason for abandoning this is there is essentially no difference between 90% and 100% modulation in the received signal sound and I was trying to minimize the number of modifications.

I did replace Q3 with a 2N3904 as mentioned in my linked post but this only works if the power input is reduced to 100 mW to avoid over heating. This change gave a higher efficiency than the original MPS3055 at 100 mW input.

The efficiency gained by changing the output filter network was worth the effort and I subsequently found by experimenting with another transmitter that T50-2 toroids give about a 10% advantage in efficiency compared to the T44-15 toroids used in the Ramsey circuit. I did not try these with the AM-25 and if you do the number of turns will have to be changed to get the proper L.

Food for thought.

Neil

hi neil,

i did some more tests yesterday. i was able to get a hold of the transistors and and diodes mentioned that tesla article. right now i have i have am using a 2n3055 as Q9 and a nte2396 as Q3. i also have diodes substituted for R12 and R33/34. this setup definitely will not pass part 15 but my intention is purely for experiments. other than very brief range tests, i don't plan on broadcasting with this transmitter.

i wrote up a summary of my experience with this setup but i'm not sure if it is okay to talk about it stuff on this website. is it okay to write about experiments that don't fall within part15?

without revealing any details i can say that my amt3000 seemed to perform better then a modified am25 despite the am25 violating part15.

hi neil,

i hope my last post wasn't problematic. my ultimate goal is part 15 radio but i am trying to do as many experiments as i can to learn about this technology.

right now i'm trying to understand how you measure modulation on an oscilloscope. you mentioned in another thread ( http://www.part15.us/index.php?q=node/571 ) that you use "trapezoidal display" using an oscilloscope that has x and y inputs. my scope has x and y inputs but i am having a hard time figuring out how to use them properly. theres not much information online.

this website " http://www.tpub.com/neets/book21/90a.htm " is the best explanation i have found but even it doesn't give me a very clear understanding. i have my antenna output hooked up to the y input and R33 hooked to the X input. is that correct?

this gives me two wave forms on my scope. i found a 'horizontal mode' setting i can change to 'XY' but that just gives me a weird blurred shape in the lower right of my display. here is a junky image of what i am describing:

any explanation would be appreciated.

I don't think your post was "problematic". This site is to encourage part 15 operation and not "pirate radio". Posting technical information should not be a problem if the major emphasis is not on illegal operation. Use your judgment on this.

The article you linked in your last post is a fair explanation of the trapezoidal display. I obtained the waveforms a bit differently than you or the article describe but I don't see why yours didn't work. Was there modulation applied? It is also quite possible that RF was being coupled into the audio scope probe, especially if an antenna rather than a dummy load is used. Make sure both probe grounds are connected to the circuit ground.

Here's how I obtained the trapezoidal display. The scope was set for AC coupling and X-Y mode. With no signals center the dot using the X and Y position controls (keep the Intensity down so as not to burn the screen). The Y probe is connected across the dummy load at the output and the X probe is connected to the audio input of the AM-25. This X connection differs from the usual method of picking up the audio at the final amp but it works with the AM-25 since the audio circuit (U1, etc.) introduces no phase shift over most of the audio spectrum. It also lessens the chance of getting RF into the audio probe. With modulation (I used a 440 Hz sine wave from a signal generator) adjust the X and Y gains on the scope to get a usable display. If the transmitter is operating linearly the sides of the triangle will be straight as shown in the waveform I posted.

The modulation percentage can also be measured with a single channel connected to the RF output and works best with a sine wave audio input. Set the sweep at about 2 or 5 milliseconds per division or to display a few cycles of the modulation envelop and sync on the audio. Most scopes will sync to the envelop of the RF display but if not then use the external sync connected to the audio input and adjust for a stable display. Call the peak deflection of the envelop A and the minimum B. The percent modulation is M = ((A-B)/(A+B)) X 100%. This equation also works with the trapezoidal display.

Either method can be used to monitor the modulation with program material but the trapezoidal is easier to interpret. Watch for "pinching off" of the carrier on audio peaks since this indicates greater than 100% modulation.

Neil

yes i was using a dummy load and yes when i modulated the signal it started looking trapezoidal. can you recommend an article or website that explains the subtleties of the trapezoid. depending on how i adjust the audio level (R31) i see patterns such as a solid squiggly vertical line on the right side of the image. also at higher frequencies i see a weird horizontal line forming in the center of the pattern.

heres some images that indicate this. sorry the quality is bad, i had to use my phone. the patterns are more pronounced in real life than in the photos. both images are a 1V Pk-Pk sine wave input. the first image is at 400hz and the second one is at 4khz.

are these just signs of over modulation?

Unfortunately, the waveforms you posted show severe audio distortion. Overmodulation would appear as a tail at the horizontal pointed end of the pattern but your patterns show nonlinearity.

It might be helpful if you apply the sine wave at the input and view the waveform at the emitter of Q9. It should be a sine wave about 8 times the voltage at the wiper of R31 (going by memory on this). A sine wave should also appear at pin 1 of U1. There will be a DC component present which varies with R23 setting.

One of the best treatments of trapezoidal patterns is in the ARRL Handbooks. If I find a good link (haven't looked yet) on the net I'll let you know.

Neil

how can you tell it is distorted? is it the overall shape? or the patterns within the shape i was describing to you?

does this image look better?

i have the 2011 arrl handbook but i don't see anything about trapezoids or xy mode in the index.