Don't feel bad AT ALL 3 MADE, I blew my first LDMOS device but mine was with ESD LOL, Since then I have my hand wrapped and grounded to the house but I see Sharky right here holding that bad boy with no problems, I am sure Im going to the extent having my dang hand wrapped in a copper grounded to the house lol lol lol, but I was like, I'm not wasting another 200 dollars ever again, Thanks MudDuckSharky for taking care of my buddy 3Made! Can't wait to try out one of your kits in the near future. - Gatekeeper

Great video ! So many people have blown up $200 transistors, simply due to tiny errors. On the Temperature switch: better to run your 1300W-capable transistors at ~800W and enjoy the Comfort factor. The extra 500W of output would only get you a fraction of an S-point on the receiver's end anyway.

Thats why i put the on off switch in back .to cut off 65volt to set bais.. my bad i did ask questions along the way but shit happens it is my first new build . I never keyed the amp .

Great video Sharky, your copper heat spreader CNC design is awesome.!

thank you !!

The bias should be limited to max gate volts so that even with a shorted pot it cant saturate the device. Iv nervously used normal solder to solder down as per W6PQLs video. Used an old stove element to heat the copper.

Good to see you back at it again.

When you refrigerate the paste it stay bound togather better. When left out the flux will dry out 😉 ps great to see your playing with Rf Again. I was deep into the water cooled pallets, I can’t wait till its main stream. If I could get my hands on a h20 cooled 8877 I would make my alpha silent ☺️nothing better then a quiet amp. My buddy just built a 4x 813 open chassis no fan beauty ☺️

Good to see ya sharky

The key to designing any analog amplifier is maintaining the allowable loss of the power device. Start by measuring the efficiency of the designed amplifier and check the allowable loss of the selected device. For LDMOS devices utilized, it is generally indicated as 1300W~1800W. You can't use all of this~ It is also important to note that the heat dissipation area of ​​these LDMOS devices is too small to allow for such losses. For 65% efficiency, the limit is 600W of continuous output. In the case of SSB mode, the average value is only about 40%, so in theory it is possible to produce about 2.5 times more output, but in reality, it is unreasonable to exceed 1.5 times as efficiency is lower. Please don't waste expensive devices! And the important thing is negative feedback. Less negative feedback causes oscillation and impairs linearity. If negative feedback is reduced due to the desire for large output, the linearity of the output decreases and becomes unstable. It is recommended to use a negative feedback resistor that covers large power. This serves to expand the BAND WIDTH of the output because it provides not only negative feedback but also a resistive load. Good luck~ From SOUTH KOREA~

Its great to see you back with another video. Enjoyed it greatly.

very insiteful troubleshooting, as I'm learning how to build my own HF amp.. this level of explanation on fixing issues is helping me avoid issues.

Possible that pot wiper went loose and depending how you have that voltage divider configured it may have passed high voltage toward the bias and saturated the fet. Or the guy went tweaking it while it oscillated. Also I don't get why CB builders use those giant TRX relays. There are RZL3-1C4 series 16A 250Vac relays for $1.50 that are better suited for 27Mhz RF due to much smaller size. Another flaw in this build is like you mentioned no temp protection and no high VSWR protection on the pallet.

Hay 3made that is a really good looking box.

I enjoy your videos...

I believe your theory is correct. There will be a big spike as the D-G feedback capacitor charges.

I really enjoyed the video, with so many catastrophic situations that can happen to this amplifier, well seen! the auto oscillation thing. Regarding the problem of placing a thermocouple, or thermoswitch, it is a very good idea, I have been manufacturing amplifiers for years, and we must take into account one thing that all builders ignore, when we use copper heat sinks, together with a copper heatsink. aluminum, we create a temperature cloud at a certain coefficient in the middle of the joint of the two materials, with which the copper part thermally collapses and stops being efficient. I manufacture my amplifiers entirely with copper heatsinks, it is more expensive, but much more efficient. Greetings from Seville Spain, EA7KNR, J.Manuel

Unlike other CB amp techs, Sharky has a deep understanding of electronics and RF theory. One other reason for not running bias during receive is that idling current will generate RF noise which may couple into receive.

Thanks for the time and effort to post this for us...Big shout out for putting family first.Lord willing there will be time to play radio besides kids learn fast you may have your very on assembly team soon hihi...take care neighbor..

It’s nice to see you making a video

Awesome thanks for working on this for 3

Love this guy legend

Thanks for posting this.

welcome back❤

welcome back GOAT

Yep great point about the high temp circuit.....

Hay mds i love your videos please get bCk to doin more builds . Your ldmos work is thee best. I hope t get one of your amps at some point 2 of those would make ne very happy 199 florida georgia line.

Thank you for this post. I learned a lot and bought one of those transistor analyzers. I agree about the thermal safeguards. I am just starting to work with LDMOS devices. Is this product made for 27 MHz? Don't see a switchable filter board.

nice fix . that a un reg supply. 240 V better but still will sag a bit. I use a NC thermal switch to shut it down . drop relay and bias. I like to cut off early to give safe room for the transistor what ever type. BJT or FET . adding more filter cap on the 65V supply can hep a bit but the power on surge may damage rectifiers. need soft start. I do not use auto key my self. PTT from radio. much better for SSB and saves relay sparking. BBI did a great video on RF flash on relay . I use IC 7300 for my radio. for LD MOS bigger attenuator. radio has PTT and ALC . hope this has some good fans . key down time ? 73's

@Sharky; did you add the resistor on the keying circuit to raise the base voltage because the Darlington transistor configurations have higher VBE? [EDIT] BTW, great to see you again!

I would like to see the voltage shape on both drains with a scope.

👍

Most alcohol isn't 100% proof. A lot of what you buy from the average store has up to 50% water in it so if the circuit was drenched in the stuff and powered up then it wouldn't surprise me the transistor shorted out. Then again I'm only 5 minutes in.

My solder paste tends to do that after about two to three years lol.

Hey man. MUch respect for what your doing. I have a question. I’ve had the tin based solder grow whiskers over time somehow and short stuff out. Where as silver lead based solder always has done what I’ve ask of it. Have u ever seen this. I’ve never seen it on a component joint. More on where wires are soldered. Was jus curious if you have ever seen it or heard about it

I have a 100 amp DC to DC power supply runs up to 54 v if I could get one of these amps what box would I need. And how many transistors would be in it. And what would be the wattage? Guesstimation?

What kind of mosfet is that style?

I know that green stuff, its due to a very agresive flux, it cleans with alcohol, but it all goes to the lowest part of the thing youre cleaning and corrodes all the copper on sight

What model NXP is this?

If you don't mind, turn on the video with subtitles.

Those huge blocks of copper are EXPENSIVE. SIR, How do you know all this? you need to be a electronic genius to make all this work.

When dealing with high voltage and or high frequency, always clean the flux. Why use so much thermale conductive paste? Make sure the heat sink surface is flat, really flat en use a very thin film of the paste.

Are you selling pallets yet?

I´ve seen some video´s from the NXP labs where they never solder this transistors. They use screws and a special clamp. I think the recommended way to mount this transistors is the best.

Are you back to selling these in kit form as you have in the past by chance?

Very good explanation. Rf can be a bitch at times. I think i would be more comfortable adjusting a linear with 4 813"s and one hand in the pocket. Keep up with the videos maybe an old dog can learn new tricks.

Sure miss your videos. Your systematic trouble shooting sure teaches us a lot!

Never put a amp clamp on.for one, a amp clamp for that low current level is not easy to find.always fuse and inline a ampmeter that way that cant happen to gate fet on.

When he said he used acetone to clean the alcohol with, I was like oh no 😮. I learned the hard way never use acetone to clean PCBs because it introduces resistance between traces and components, it messes up your whole circuit. I had to remove all surface mount components to clean the acetone off with alcohol to fix my PCB. So be careful and avoid using acetone to clean your PCBs.

I think "Eutectic" is the word you are looking for.🙂 63/37 eutectic point is 183 degrees C

The balls 3made must have to have is work looked at by one of youtubes ldmos experts.

I haven’t gotten through the entire video yet, but isn’t that transistor package made or beryllium oxide? The dust of which is extremely carcinogenic. The NXP LDMOS final stage parts that I used to work with are. And the vendor would warn us repeatedly not handle a fried part with its package blown off. A blowtorch wouldn’t help the situation, i guess.

S

27 MHZ ? Give me a break ! Very Illegal !

WELL GUESS WHAT FELLOWS. the FCC has woke up and is now watching what you are doing, what you are running and what kind of trouble you are causing, the last fine I saw was 25,ooo$. Have fun.