Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: Prev 1 [2] 3 Next   Go Down
  Print  
Author Topic: Amp Build Project - is the GS-35B a good or bad choice?  (Read 12845 times)
K4FX
Member

Posts: 166


WWW

Ignore
« Reply #15 on: September 14, 2011, 03:31:39 PM »

Jeff.

Those 8877's you were mentioning were from the late 80's I think it was. That was a manufacturing problem, but the ones since are great tubes just as Bill mentioned. I have a 1997 vintage that makes 1500w with a little less than 50w in. It's a solid tube.

The thing I like about 8877's is you have a lot of purchasing options, Chinese and domestic new, and many full output medical pulls for 350 or so bucks.

The 3 minute warmup does seem like an eternity. That's why I mainly use my 8877 in contests, and I have an old SB-220 I use for  it's instant on "when ever I need it" power.

Good luck on your project,

K4FX

PS It seems like one of the 3CX3000's has more plate dissipation than the others, I think it is the /F7 isn't it?
Logged
N3JBH
Member

Posts: 2358




Ignore
« Reply #16 on: September 14, 2011, 11:50:48 PM »

i think the plate dissipation is same on them. F7 has leads on it that makes it  not as good for a amp when your getting in to the higher frequencies but not a expert by a million miles on that stuff. Hoping one would have chimed in.
Logged
KC2RLY
Member

Posts: 37




Ignore
« Reply #17 on: September 15, 2011, 11:10:14 AM »

I think the 35b is an OK choice but having built a few here are some tips:

It is an OK tube, maybe a bit less efficient than others but who cares. This isn't Nasa's deep space probe and this isn't rocket surgery. Its a homebrew multi band amp for ham radio.

It is happy with higher voltages so use them. the lower plate impedences are easier to match and it lives quite happy from 3500-4200v.

I recommend a minimum supply of 750ma ccs @2800vac, after the rectifier and caps you will be around 3900-4000v. I typically use 1a ccs plate transformers.

Bias Bias Bias, the w4zt board is an easy option but if you want a brute force solution you can feed -34v to the k via a choke (the larger ring)

YOU CANNOT use a CT filament transformer to supply bias.

You CAN use a couple of radio shack transformers for the filament. 12.6-13.9v have been used with no problems. 1 tube=two RS transformers, 2 tubes= 3 transformers.

If I am not mistaken the last one I built for 6 meters needed about 100w pep to do about 1400-1600w pep out. this was at 3900v.

Your mileage may vary from tube to tube. keep this in mind when running multiple tubes in parallel.

A parasitic choke is a requirement because it CAN amplify way into vhf and beyond. See W8JI's site (or maybe it was Rich Measures who gives specs on how to build them).

The price has been going up for years on these. Right now they aren't cheap but are not too expensive. Since these are SURPLUS, buy your spares now before they go the way of the sweep tube and you'll be fine.


Logged
W6WRT
Member

Posts: 26




Ignore
« Reply #18 on: September 15, 2011, 01:18:25 PM »

A parasitic choke is a requirement because it CAN amplify way into vhf and beyond.

I have built two GS-35b amps and a number of others using 8877s, 4CX1000s and 4CX1500sand none of them needed a VHF parasitic suppressor. The secret is to reduce the gain of the tube below 1 at the VHF frequency. Do that and it will not oscillate.

The traditional way of reducing the gain is to use a VHF parasitic suppressor of course, but there are other ways too. What I have done is to use s relatively high Q input circuit (Q=5 or more) and keep the leads in the input circuit short and very close to the tube. This reduces the VHF impedance in the cathode circuit down so much that any VHF energy coupled back into the cathode circuit is swamped out by the low impedance and therefore the VHF gain falls below 1. All of the amps I have built this way have been absolutely stable, even with nothing connected to the input or output, or with any combination of bands and loads.

My advice applies only to grounded grid amps. I don't do grid driven amps.

The major advantage of eliminating the traditional VHF suppressor is you also eliminate the juggling of inductor and resistor values in order to find something that will not burn up the resistor while operating on 10 meters. Anyone who has homebrewed their own design will know what I am talking about. Give it a try - eliminating that nuisance suppressor is well worth the effort, IMO.

Bill, W6WRT
Logged
KO7I
Member

Posts: 106




Ignore
« Reply #19 on: September 15, 2011, 10:08:40 PM »

I am looking at the GS-35B dot com web site data sheet on the tube. I am reading that the Cathode current is only 1.4A. Am I reading that correct?
Greg, KC2RLY, When you say the "K" ring, the way I read the schematic of the tube layout you are refering to putting -34Vdc in the cothode. I presume you are grounding the "C" ring which is the grid. Correct?
Assuming 3500V Anode voltage, with -34Vdc what is the expected bias current?
Thanks & 73, Don KO7i
Logged
KC2RLY
Member

Posts: 37




Ignore
« Reply #20 on: September 16, 2011, 08:52:35 AM »

I am looking at the GS-35B dot com web site data sheet on the tube. I am reading that the Cathode current is only 1.4A. Am I reading that correct?
Greg, KC2RLY, When you say the "K" ring, the way I read the schematic of the tube layout you are refering to putting -34Vdc in the cothode. I presume you are grounding the "C" ring which is the grid. Correct?
Assuming 3500V Anode voltage, with -34Vdc what is the expected bias current?
Thanks & 73, Don KO7i

I think the max current measured at the cathode (proper way) might be 1.4 amps but thats alot. Most I have never had past 1 amp plate current

When I say "K" ring I meant cathode which is the larger ring. On the ruskie datasheets its shown labeled as "K".

Yes I direct ground the Grid.

No, I put -34vdc to the cathode through a choke just like a plate choke. I don't honestly know what the bias supply current is, I cant imagine its much. I use 6a10 diodes for everything from 3cx6000a7's down to 4cx250b's.
Logged
KO7I
Member

Posts: 106




Ignore
« Reply #21 on: September 16, 2011, 11:55:15 AM »

My bad.
I meant to ask you what the idle current was when enabled but no drive.
Thanks, Don
Logged
K4FX
Member

Posts: 166


WWW

Ignore
« Reply #22 on: September 16, 2011, 03:38:10 PM »

I checked the eimac web site, according to it, the 3cx3000/f7 is 4000w plate dissipation

I thought it was more than that but i was wrong.
Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #23 on: September 16, 2011, 04:07:55 PM »

I checked the eimac web site, according to it, the 3cx3000/f7 is 4000w plate dissipation

I thought it was more than that but i was wrong.

I designed several commercial 3CX3000 amps, and long ago even built some Ham amps. Eimac made a custom flanged grid for one amp I designed, with flying leads for the filament. This allowed the tube to removed from the top side like a normal socketed tube, but without a socket.

All of the 3CX3000 series and even the 3CX2500 tubes had 4000 watt coolers. They can be even better than that with more pressure drop across the cooler, if you watch seal temperatures, but they start to choke up pretty fast.

They are instant on. Good tubes. The -7 high mu series are exceptionaly clean on SSB.

73 Tom

Logged
K9FV
Member

Posts: 479




Ignore
« Reply #24 on: September 17, 2011, 09:06:24 AM »

My bad.
I meant to ask you what the idle current was when enabled but no drive.
Thanks, Don

Don, I just finished a 6 meter amp using a GS31b tube which seems to be the same as the GS35b tube, except derated for lower output and longer life.  I am using a 30 volt, 50 watt zener in the bias circuit which gives 80mA idle current with 3200 vdc plate voltage.

Hope this helps,

73 de Ken H>
Logged
KO7I
Member

Posts: 106




Ignore
« Reply #25 on: September 21, 2011, 07:28:29 AM »

Thanks Ken,
One last question (I hope HI HI). At full power out, what was/is the grid current? I am picking out meters and want to make sure I scale them correctly. I will have 3 meters, 5kV Plate V, 2A Plate I, and thinking 300mA for Grid I.
I guess a better question is what should the grid current for a GS-35B be limited to?
Thanks, Don
Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #26 on: September 21, 2011, 12:25:13 PM »

I'm not reading through all of this, but of the GS35 is an oxide cathode it almost certainly should have a warm up timer and a grid overload disconnect.

As for stability, with a properly grounded ring is should be unconditionally stable:

http://www.w8ji.com/vhf_stability.htm

It has nothing to do with gain<1 at VHF.  The most stable tubes are actually tubes that also work best at VHF. Stability is all about grid resonance and plate impedance.

As for metering:

http://www.w8ji.com/metering_amplifier.htm

With an oxide cathode emitter, be sure the heater voltage is centered on the specified nominal heater voltage at the tube pins. Be especially sure it never goes outside low range. Tubes fail much more rapidly outside low heater voltage limits than the do at higher than limit voltage. 

Logged
W6WRT
Member

Posts: 26




Ignore
« Reply #27 on: September 22, 2011, 11:34:13 AM »


It has nothing to do with gain<1 at VHF.  The most stable tubes are actually tubes that also work best at VHF. Stability is all about grid resonance and plate impedance.


Nothing to do with gain<1 at VHF?  Perhaps I wasn't clear in my original post. I was talking about the entire circuit, not just the tube. If the gain of the entire circuit is less than 1, it will not oscillate. As we are all aware, overall circuit gain is made up of many factors, not just the gain of the tube alone.

Many builders make the mistake of ignoring the grid-cathode impedance at VHF. By having it unnecessarily high, VHF stability is decreased. Keep it low enough and the fed-back energy will be swamped out, the gain of the overall circuit falls below 1 and the circuit will not oscillate.

Bill, W6WRT
Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #28 on: September 22, 2011, 12:29:01 PM »


It has nothing to do with gain<1 at VHF.  The most stable tubes are actually tubes that also work best at VHF. Stability is all about grid resonance and plate impedance.


Nothing to do with gain<1 at VHF?  Perhaps I wasn't clear in my original post. I was talking about the entire circuit, not just the tube. If the gain of the entire circuit is less than 1, it will not oscillate. As we are all aware, overall circuit gain is made up of many factors, not just the gain of the tube alone.

Many builders make the mistake of ignoring the grid-cathode impedance at VHF. By having it unnecessarily high, VHF stability is decreased. Keep it low enough and the fed-back energy will be swamped out, the gain of the overall circuit falls below 1 and the circuit will not oscillate.

Bill, W6WRT


Hi Bill,

The normal mode of oscillation is TPTG mode, where the grid is parallel resonant. This is true in almost everything except perhaps tubes with high plate-cathode coupling used in grounded grid, where they oscillate near the operating frequency.

What the circuit does outside the tube doesn't matter much, except the anode. The grid is a circuit inside the tube onto itself.

The cone grid connection in a ringed grid terminal, like a 3CX800A7, 8877, or 3CX1200Z7 makes the tube unconditionally stable without suppression if the grid ring is properly grounded. The same is true for a 3CX3000A7.

So while it is the gain of the "circuit", we have to be clear the "circuit" is almost always inside the tube and not outside.  :-)

People might confuse gain of the circuit as meaning at the normal input and output, or in grounded grid the cathode to anode gain, when it is really almost always the anode to grid gain even in grounded grid.


73 Tom
Logged
K4EJQ
Member

Posts: 100




Ignore
« Reply #29 on: September 27, 2011, 09:14:52 PM »

Hello All; I am just finishing up a new 160 through 6 meter amplifier using the Russian GS35B tube. The project has not been without some set backs from time to time but it has been a real "learning experience" for this OM. I have been building amps for nearly 55 years now and each one has been an experience.

 The plate efficiency of this latest amp varies of course with frequency: from about 55 percent at 6 meters up to nearly 68 percent on 17, 20, 30 and 40 meters. The efficiency then begins to roll off to around 63 percent at 80 meters and down to about 55% at 160 meters. I spent too much time and money trying to improve the performance at 160 meters. Several different tank circuit parameters were tried on that band with little improvement noted. I am driving the amp with a little IC-706 transceiver. The 706 has only 70 watts output at 50 Mhz so that might explain the 55 % efficiency I'm getting there. The amp has a tuned input.

I built the amplifier primarily for 6 and 160 meter operation-CW. It runs 1 KW output on those bands and a good bit more on the other bands. Plate voltage under load is 3200 volts. Using a simple quarter-wave shorted stub on the output at 6 meters has the second harmonic (FM band) down almost 70 dbc. Tuning is smooth across the bands. Grid drive peaks at or very near plate "dip" with PO peaking at same. Amp is stable with input and outputs unterminated with bias set for an idle current( NO drive applied) of 125 ma of plate current. Plate tuning was run through it's entire range on each band-NO sign of oscillation viewing a spectrum analyzer coupled to the transmission line connected to the output port which was unterminated.

Power gain of the tube measures about 13 db. A good spare tube was obtained for less than $90 shipped. Cost to build was well over $1K even though the plate transformer from a 1KW AM broadcast transmitter was a freebie. Would I recommend the GS35b for a homebrewed amp? Yes, what with the costs of all vacuum tube amps skyrocketing. Just stock up on the type tubes you use before they become unavailable.

So far, so good. YMMV. 73, Bunky, K4EJQ
« Last Edit: September 28, 2011, 05:12:27 PM by K4EJQ » Logged
Pages: Prev 1 [2] 3 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!