BTI LK-2000 Amps - Huh?

[N8ETQ]

BTI LK-2000 Amps


   Brutal.... I got 2 amps, 3 schematics, Nothing Jives! I got them
both running. One is sporting a 3CX1200, the other a 4X1. Both have
4kv on the plate. One is 110, other is 220. Aye Vey.... My Cantenna
Needs a day off! I could use a cocktail myself. Still, I wish they were
mine!

/Dan

http://www.eham.net/reviews/detail/2680

[kb3ouk]

Somebody must've done some tube changing, thought they had a 3-1000 in them?

[N8ETQ]

  Yeah,

       From what I can figure out the standard model came with
a 3X1, If you wanted you could get them to ship it with a 4X1.
I guess if you were already sitting on a 4X1 amp and a room full of
spares. Either way the 3CX1200 fits in the same socket.. The screen on 4X1 screen is grounded as well as the grid.

/Dan

Somebody must've done some tube changing, thought they had a 3-1000 in them?

[Opcom]

The 4-1000 loves high voltages in GG. Remember the Elenco Commander?

Having a chassis for a 3-1000 leaves open the possibility of a 4-1000 as long as the extra 1.75" or so height requirement of the latter is considered.

For the inverse purpose, the metal fingers that usually are installed to contact the 4-1000's base cooling cylinder must not be so tall as to interfere w/ the 3-1000's envelope. Because of that cylinder which is really part of the 4-1000's base seal cooling system, the air system socket is best for the 4-1000, even if people tend to dislike it.

Do you have good pics of the insides of the amps? It would be very interesting to see how each tube and socket is mounted and the cooling. I've never seen a 3CX1200 in-socket.

Did I misunderstand the comparison between the 3CX1200 and 3-1000 basing?

4-1000 - chimney  SK-506 - socket SK-500 - circle dia. 1.5",  pins ~0.376
3-1000 - chimney SK-516 -  socket SK-510 - circle dia. 1.5",  pins ~0.376
3CX1200A7 - chimney SK-436 -  socket SK-410 - circle dia 1.25", pins ~0.190
3CX1200D7 - chimney SK-446 -  socket SK-410 - circle dia 1.25", pins ~0.190
 (this is a 6.3V tube and 0.4" shorter than the A7)

[N8ETQ]

Hey Patrick,

     According to the instructions it just required a hack saw..

Classic!

/Dan

[WA3VJB]

     According to the instructions it just required a hack saw..

That's great!!!

I just went through an ordeal getting a firmware "upgrade" pushed to my smartphone.  I thought about a hacksaw at times, but it wouldn't have been an upgrade, huh-HA !

[N8ETQ]

Hey Patrick,

   Hi Res. camera's changed my life! Here's a couple Pix..

/Dan

Do you have good pics of the insides of the amps? It would be very interesting to see how each tube and socket is mounted and the cooling. I've never seen a 3CX1200 in-socket.

[K1JJ]

The BTI sure is a nice ham amp.  I had one myself at one time.

If I had a choice, I would prefer the 3CX-1200  or the 3-1000Z  simply because they were designed for linear service.  They will show the best intermod products as good zero-bias power grid triodes do. I'd expect maybe 5-7db better 3rd order than the 4X1 in GG.

The 4-1000A tetrode will work OK as a GG linear, as many hams have shown. But the triodes in GG will be cleaner. This can make a big difference when ssb stations are close together and trying to operate within 3kc spacing.   On wider band AM, the same "relative" bandwidth degradation will occur, since we are still mixing frequencies, but at higher tones.

So, bottom line, if we have a choice, go with the triodes. If not, then probably no one will say anything anyway... 

T

[N8ETQ]

Hey Tom,


   Good point, Agreed, tetrodes have IMD issues..
OTOH, I just fix em'. and believe me, these were
FUBAR....

73

/Dan

The BTI sure is a nice ham amp.  I had one myself at one time.

If I had a choice, I would prefer the 3CX-1200  or the 3-1000Z  simply because they were designed for linear service.  They will show the best intermod products as good zero-bias power grid triodes do. I'd expect maybe 5-7db better 3rd order than the 4X1 in GG.

T

[K5UJ]

what is it about the additional grid when both are grounded that causes more IMD?

tnx

rob

[K1JJ]

what is it about the additional grid when both are grounded that causes more IMD?
tnx
rob

Hi Rob,


Background:

Both the 4-1000A and 813 appear to be anomalies, in that they do very well in linear service when triode-connected. In fact I had a pair of 4X1's triode connected as modulators and they sounded FB. I have since changed them over to grid driven/ screen voltage conventional tetrode modulator service and they do even better, simply cuz they were designed for this. Using regulated supplies and the FET solid state driver, they perform excellent.

If an audio amplifier exhibits -30db 3rd order IMD, there is no way we can hear any distortion with the human ear. However, an RF amplifier on the band that is close by will easily show the difference between say, -30 to -40 db IMD by the spaltter products. It is easy to see the difference.  



Tetrode in GG  vs: Power Grid Triode in GG:

That said, when a tetrode is placed in GG service, it gets an additional boost in cleanliness due to the inherent negative feedback of GG. And since the design already works well in triode-connected service, it's all the better.   But alas, the 4X1 was not designed or optimized to be run this way even though it still performs well. The engineers did not FINE-TUNE element spacing, internal structures and operating parameters to give the "perfect" plate transfer curves of a linear amp in GG or simply triode-connected.  Instead, they were optimized for tetrode / grid driven service.

But in contrast, the 3-1000Z and family of external anode power grid triodes WERE optimized for GG service. Thus, they have that cleanliness edge when run the way as intended, in GG linear.

Again, a 4X1 in GG is a pretty clean stage. I have run them and have had no splatter reports. It seems a few who run them in grid driven tetrode RF linear service have problems simply from unregulated screen and/or grid voltages, improper loading, overdriven grids or the multitude of instability problems of a high gain stage.

In comparison,  a GG 4X1 is pretty clean as long as loaded heavily - is easy to tame and quite a docile beast, especially on the higher bands.  But if we have a choice, go with a specialty triode designed for the most linear performance. Most simple and effective.

T

[K5UJ]

Thanks Tom.  I wonder if the 4x1 IMD spec. in gg improves with an increase in Ep.   I think I read somewhere that it actually gets better with a decrease in Ep with the 3-500Z.  Don't know about the 4-1000 though. 
 

73

Rob

[K1JJ]

Thanks Tom.  I wonder if the 4x1 IMD spec. in gg improves with an increase in Ep.   I think I read somewhere that it actually gets better with a decrease in Ep with the 3-500Z.  Don't know about the 4-1000 though.  
 73
Rob

Rob,

I guess we could always get out a ruler and compare the straightness of the transfer curves at lower vs: higher plate voltages to see the effects.

But, you are correct about the 3-500Z finding a sweet spot at a lower plate voltage. Eimac printed a sheet showing a 3-500Z running GG at 1500V has 3rd IMD at about -40db. This is quite clean and linear. In fact, I took this to heart and have as a driver, a pair of 3-500Z's running at about 1500V that are loaded extremely heavily. The loading is so heavy that the C1 plate tuning peak is very broad. But the measured IMD performance (using an SDR receiver spec analyzer) is quite spectacular. It only puts out about 300 watts pep, but what a nice 300 watts.  It's the cleanest amplifier I have in the shack.

Now, for the 4-1000A, I understand that 5500 VDC is the maximum anyone should run the tube in linear service. This is because of extreme electron flow bouncing off the plate structure creating lost emission, non-linear -  or something like this.  

I have never seen anything about running the 4X1 at reduced  voltage like the 3-500Z, but since the 3-500Z has been proven to act this well at 1500V, as a driver or "lab amp," why not stick with a pair and be done with it?  

The only downside is the 3-500Z is expensive and in high demand, whereas the 4X1 is relatively easy to find.  I'd be very intersted in seeing the IMD results of someone running a 4X1 in GG with about 1800V and loaded very heavily. It would probably require a positive grid-to-fil bias to idle high enuff, but this could be done.



T

[K5UJ]

Okay Tom FB.  at 1.5KV I wonder if the 3-500Z grid current is high.  I note on the 4x1 triode Eimac memorandum data sheet Ig goes down with an increase in Ep which is understandable.  I was assuming they don't recommend less than 3 KV on the plate because the grid current will be too high although I have not looked at the data sheet to see what the 4x1 max Ig is. 

I've had good luck finding nos reasonably priced 3-500Zs and even 3-1000Zs (although I passed on them) but all the 4x1s I've found have been duds or low output pulls.

I guess I hijacked the thread.  sorry about that. 

[K1JJ]

Rob,

When running 1500V on the 3-500Z's, we intentionally run low drive for cleanliness and monitor the IMD. If we drew excessive grid current, it wud probably show up as poor IMD.  After all, it's a driver / "barefoot" amp - not used as a kick ass final-final.

I would say our discussion is right on topic, being these are the tubes used in the BTI...

Well, off to do some sailing.  Nice wind today!

T

[KM1H]

The 4X1 in GG service has better IMD than a lot of the amps currently on the market. So does the 813, 4-400A/4-250A, PL-175A and a few others.

In AB1 its good for around -30 3rd order which is so-so. The biggest problem is getting low impedance regulators, zeners and VR tubes dont cut it. Something like G3SEK's tetrode boards are needed and can easily add 3-5 dB more improvement.

In GG add another -5 to 6dB and its right up there with the better ones.

That tube got a bad rap when having only 2500V on the plate and ramming it with 125+ W of drive from sweep tube rigs. GIGO
Put some real voltage on it and chill out

Carl

[K1JJ]

I guess we're all preaching to the choir...

Here's a set of quad 4X1's in linear grounded grid.  This was a fun project. It ran like a champ.  

I later converted it into a single class C final, plate modulated by a pair -named "Fabio."  Still running strong today.


T

[ke7trp]

The BTI is one cool vintage Amp with the 3-1000.  I have always wanted one. I know it was popular to reconfigure the power supply for more voltage.  I dont know details but you could change the configuration and get 4500+ volts on the plate and the amp would stomp.  They may have shipped it tapped at a low primary for input regs???  Maybe changed to a full wave?    Cap input?

They where also made by a second company. I have old adds that list the same amp, features and even model number but where Halfstrom?  Halstrom?  Something like that.  I think BTI must have sold the rights off?  

We had a harris amp that used a 3-1000.   Looked like a dishwasher.  Never tried to get it running.  Sold it to an AM'r and hope to hear it one day.  

C

[Pete, WA2CWA]

BTI was Brad Thompson Industries; the amplifier later was manufactured by Hafstrom Technical Products.

[ke7trp]

Yeah.  Thats it Pete.  Hafstrom.  Neat amp for a ranger or vintage SSB.

[N8ETQ]

Hello All,

   I can't be sure as the documentation on these
things is a mess, but I will offer what I "Think"
I know.

   The original BTI LK2000 Amp (Schematic dates
8-66,10-66,12-66) Had an "Antenna Load" Sw. on the
Front panel for the optional Dummy load "Dl-2000".
There is an "ALC" adjustment pot on the P.Sup.

   Then it seems there was a Variation called the
BTI LK2000 HD. "Heavy Duty". As far as I can tell
they added 2 additional filter Caps in the HV supply
with associated bleeders to allow the Voltage to rise
a bit.

    Next is the Hafstrom BTI LK2000 that can be
identified by the "ALC" pot being moved to the
front panel in place of the "Ant Load" Sw. Also the
Jones plug for the "DL-2000" was changed from 6 pin
to 12 pin. Indicating the later model may not have
supported the DL-2000. Schematic dated July 1971.
Serial numbers above 3000 were probably Hafstrom.

   Here is a pix of the 2 side by side. The one on
the left is an original BTI but has a "Hammy" HD
Mod. The one on the right is a Hafstrom. Both have
replacement plate transformers. Both of these are
working (Now) but still need some "Clean up". Had
to get em off my bench. (No easy Task!)

73

/Dan

[N8ETQ]

 BTW, Here is a pix of a stock BTI Plate xfmr
with a HB diode/Cap Board.

/Dan

[ke7trp]

Cool old amps.  What was the power output once repaired Dan?   

C

[N8ETQ]

Hey C,

      Hard to say, the one on the left with the "HD"
mod has a 3CX1200 in it and really cooks. 1500W OUT
on 40m, Can't run it on 75 due to blown loading padders.

   The one on the right has a 4X1 in it, with a 110 input
plate xfmr. Rest of the p.sup is crap, just cobbled up with
junk box stuff. The owner is going to bring me some "Stuff"
but we'll see.  As it sits it will make 800W OUT on 75
and 40m.

    The 4X1 needs quite a bit more drive, 80W as compared
to the 3CX1200 who seems pretty happy with about 50 W.
The primary side of the 4X1 plate xfmr is pulling around 30
amps and it sags pretty good doing it. Ep is around 4KV
but once driven the 4X1 sags to <3...

   But your right! They are Cool old Amps!

73

/Dan

[ke7trp]

I would like to have a factory one.  I have 220 in here. Or should I say 246volts.  I would use it with a ranger most likely. But, I am not searching one out.  If I find one, cool.  If not, Cool. 

I had an amp with a 4pr1000.  It was very well made with Vac variables, All custom, aluminum chassis. The RF deck had a huge tank, band switched, vac relays ect..  It had a massive PWD transformer/PS on the floor on castor wheels.  The HV was 6900 volts on the plate sag to 6700 with 100 watts Drive.  MAN it smoked. I never ran it on the air as I had no antenna or tuner that would take that power.  But we did run it into the dummy a few times before it was sold.