4 x HG 2879 advice

[26OD004]

Hi all,

A friend of mine recently imported a 4 pill amp from a builder in the US and has managed to pop pill number 1 and blow the base emitter 10 ohm resistor. He contacted the box builder and was told that replacing pills 1 and 2 with a matched pair and replacing the BE resistor should be all that was needed.
I did the work 2 weeks ago and now the amp is back with me because pill number 3 and it's BE resistor are now blown, plus the balance resistor on the output combiner is smoked. Luckily he bought a matched quad so I will be replacing pills 3 and 4 over the weekend.

He was told the amp would be good for 100W input (SSB) and should expect over 1KW out, he has a video from the builder showing just shy of 900W on the meter from 80W drive (with 14V supply).

I think he's overdriving it so I'm wondering what other people are driving similar amps at and what the realistic expectations are from this amp. I've tried to do some research on the HG pills but from reading on here they just copied Toshibas data sheet so it's all pretty meaningless as the HG transistors are not a direct copy and behave differently. One of BBI's youtube video's suggests 12W drive should give around 130W out per transistor, which sounds reasonable.

He's using a 100W HF radio for the drive so I'm considering fitting a 3dB attenuator to the amp input. If it was my amp I'd also want some sort of high swr / overtemp protection installing too.

Any advice on how to stop it going bang again is more than welcome!

 

[Wire Weasel]

Simply sounds like it was repeatedly overdriven .... perhaps also into a little SWR.
Getting 900W on SSB is way too much for 4 of those pills. Were those DOSY watts? ...... nevertheless he saw smoke didn't he?
.
The 2879 Clones are 100W pills. Therefore SAFE Max out is 400watts on any mode.
You can push the drive and up the voltage ..... but then HOW LONG until you see the smoke?
If you want to see closer to a KW ..... then you need an 8 pill box.

 

[Shadetree Mechanic]

I have a few different 4 pill amps, and I only give them about 50 to 60 watts input. This gives about 600w peak output at about 15 volts.

One thing that I like to do, is to increase the negative feedback to give more stable operation.

I smoked one of my 4 pills a while ago, and the members here helped me get it going again. Here is a link to the thread with the pictures showing the 100 ohm 2 watt feedback resistors doubled up.

Post in thread 'X-Force 600HD Smoked' https://www.worldwidedx.com/threads/x-force-600hd-smoked.249976/post-729109

 

[BC Coyote]

a video from the builder showing just shy of 900W on the meter from 80W drive (with 14V supply).

Seems like every amp builder out there overdrives the snot out of the amp for the "demonstration video". 900 watts from a 4 pill is ridiculous. As mentioned above, 400 watts max is more like it if you want the box to last more than a month.
People have been falling for this BS for a long long time ! Some of the claims of amp performance I hear on YouTube and on the air are comical........

 

[26OD004]

Appreciate the replies guys, you're saying exactly what I thought you would. I just watched the demo video again and it was doing 750 - 850 peak on 80W drive @ 14.8V. Sad thing is he already has a KL703 which can do the same power out and he was trying to get something that did a bit more. I guess he better save up for an 8 pill

 

[BC Coyote]

he was trying to get something that did a bit more.

The other thing people don't realize is you need to quadruple power output to get a 1 S unit increase of signal at the receive end. So if you're friends signal is hitting you at S5 with 400 watts, he would need to go to 1600 watts to hit you at S6.
1000 watts doesn't go twice as far as 500; It doesn't work that way !

 

[26OD004]

The other thing people don't realize is you need to quadruple power output to get a 1 S unit increase of signal at the receive end. So if you're friends signal is hitting you at S5 with 400 watts, he would need to go to 1600 watts to hit you at S6.
1000 watts doesn't go twice as far as 500; It doesn't work that way !

But what about the bragging rights of running big power? Surely that makes the S meter move further?

I think the sweet spot is probably around 4 -500W, after that the law of diminishing returns kicks in as you say. It would be much more effective to put more money into your antenna system at that point.

 

[linearone]

The typical hg 4 pill now, if they are the c variant, is 800w or so. Toshiba 5 pills routinely did 1-1.2kw.
Crusher made alot of posts years ago about the hg's. He found out, along with alot of others, just matching the pills, while crucial now, is not enough. Every part needs to be matched to make them stay together. Toshibas didn't care, pop 6 pills in a 16 pill, throw 6 new ones in, lay some solder in the case, shake it a few times and re key the box..and it was fine.

 

[linearone]

The typical hg 4 pill now, if they are the c variant, is 800w or so. Toshiba 5 pills routinely did 1-1.2kw.
Crusher made alot of posts years ago about the hg's. He found out, along with alot of others, just matching the pills, while crucial now, is not enough. Every part needs to be matched to make them stay together. Toshibas didn't care, pop 6 pills in a 16 pill, throw 6 new ones in, lay some solder in the case, shake it a few times and re key the box..and it was fine.

True story about a 16 pill toshiba box. I built one for my buddy a late sk, he and i were dicking around with his station, switching stuff all around etc. we finally went to re key to test and we had really poor receive. We keyed the 16 pill and my locals were saying we sounded all fucked up. They could make us out but something was wrong.
We kept keying another 2 or 3 times, Wondering why the meter looked strange.
We had the coax un hooked from the box.
We were keying a 16 pill on volts into thin air from the so239... Box still didn't blow up . Ran fine.

 

[nomadradio]

Hmmm. My ancient Kenwood TS-50 uses two 2SC2879 for the final stage, delivers 100 Watt peaks.

Original 30 year-old Toshibas still do the job.

The incentive for the amp builder is to promise as many Watts per transistor as he possibly can. Numbers are what sell. An otherwise "wise" consumer would naturally want the most power per buck, right. Nobody asks how many times they can key the mike between kabooms.

But it's not the same as buying a 25-pound sack of dog food. Wear and tear determine the life of any human-made machine. And the 2SC2879 is no exception no matter who made it.

Cutting the wattage in half, or cutting the power supply voltage by a quarter should typically multiply the service life of a transistor only rated for 120 Watts each by the designer.

You can put a turbocharger on a 2-liter 4-banger and get big horsepower. And short service life before it grenades. Or use the stock SP injection and get 200,000 miles.

It's just a tradeoff. Either performance or reliability, or some of each.

73

 

[nomadradio]

Box still didn't blow up . Ran fine.

Like the guy who fell into the outhouse and came up wrapped in tissue?

I'll guess this is a "test" you didn't repeat on purpose.

73

 

[linearone]

Like the guy who fell into the outhouse and came up wrapped in tissue?

I'll guess this is a "test" you didn't repeat on purpose.

73

Ha ha, certainly not . And I didn't tell that story as a testament to my boxes, it was just pure luck and the fact that the toshibas were rugged. I also think if the coax was shorted that it would have certainly blown up, I think having it open is what saved it

 

[BC Coyote]

if the coax was shorted that it would have certainly blown up

LOL! That's an understatement ! I'm certain there would have been some serious fireworks..........

 

[26OD004]

well I've replaced the 2nd pair of pills now and I have advised 50 - 60W max input drive. Hopefully I don't see that amp again..........

 

[Xracerx13]

what HG's were they? were they regular 2879's or 2879C's?