T368 with 4-250s

[ke7trp]

I thought I would post some data on my T368C.  I recently converted to 4-250s for the modulator. 

The 4-125s on paper will fully modulate the rig to 100%.  In reality. They never reach 100.

Testing into the bird dummy load with the audio gen produced about 70 to 75% modulation positive.  The tips on the scope would flat top anything past this.  This is fine. The T3 runs geat and has for years.   

I got some 4-250s recently and decided to install them.  You can plug these right in but the filiment transformer runs hot.  The solution is to add another transformer under the chasiss. After this, Everything runs cool.  You can remove the stock trans and replace it wtih a 30 amp unit but I like to keep things fairly stock looking. I have heard from T3 owners that did not replace the filiment trans and it ran for years with no issues.

The tubes plug in but its nice to remove the threaded rods that held the clamp system for the 4-125s.

You will have to regap the modulation transformer.  The Gap is to tight for increased modulation and it will arc across.  I just adjusted it out a 1/16th.  No more arc.

The mod current meter on the T3 is adjusted to 2.5 for the 4-125s. This is 50MA. The 4-250s require 100 to 120 MA or about 5 on the meter.  There is plenty of range on the T3 to handle the adjustment. I tested different idle currents and saw no difference what so ever and decided on 80 to 90.

Numbers before on Power master NIST calibrated meter, Tech 2245 Oscope, and Bird dummy load:

MY numbers for MY tubes and MY transmitter. On 40 meters. Your results may be different:

4-125s
1260 watts PEP at flat top.  Scope shows 70% peak. While talking normaly 1100 watts or around 60 % mod.

4-250A
1880 watts pep before flat top.  Scope is showing 120% pos.

Carrier is 400 watts at 200 MA.  500 at 275 ma.  Peak modulation does not change much  with more loading.

Now I can load to 200MA and have 400 watts of carrier and I end up with a full legal limit power out of the T3 just under 100% mod. The signal is spotless clean on the Spec AN.  60! db down. Probably because of the Dual L network. 

I know this is nothing new and this has been done for many years.  I just wanted to post my findings and data on the change. People have swapped 250s and 400s in before I was even alive on T3s.


C

[k4kyv]

I thought I would post some data on my T368C.  I recently converted to 4-250s for the modulator. 

The 4-125s on paper will fully modulate the rig to 100%.  In reality. They never reach 100.

Testing into the bird dummy load with the audio gen produced about 70 to 75% modulation positive.  The tips on the scope would flat top anything past this.  This is fine. The T3 runs great and has for years.   

I would never be happy running any transmitter that wasn't capable of at least 100% modulation positive. Preferably even more, 120-125%, enough head room in the positive direction to allow 100% negative without flat-topping on the positive.  Flat-topping on the positive generates exactly the same kind of spurious distortion products (aka "splatter") as overmodulating in the negative. Limiting the percentage of modulation to 70% would cut your total sideband power by one-half!

I am surprised you are able to achieve 120% positive using the stock modulation transformer, even with the 4-250s.  The mod xfmr in the T-368 has roughly the same impedance characteristics as the one in the BC-610.  The total primary to secondary step-down ratio is something like 1.7:1 (3:1 impedance step-down ratio).  This amount of  step-down was deliberately in the design in attempt to avoid even the possibility of overmodulating in either direction, and usually allows modulation peaks to hit the mid 90s before flat-topping occurs.  Better positive peak response could be attained using a replacement modulation transformer with less step-down, 1.4:1 to 1.5:1 turns ratio, since the modulator and final run off the same +HV supply.  Many older broadcast transmitters and communications transmitters, as well as ham transmitters like the DX-100, incorporated the 3:1 impedance step-down in the design as a crude high level peak limiter to prevent overmodulation, but this fails to take into account the  natural asymmetry of the human voice. The mod xfmr in the Gates BC1-T is approximately 1.3:1 step-down turns ratio.  This was designed after the FCC let AM broacasters  run up to 125% positive, and most of the older  transmitters didn't have the capability.

[ke7trp]

Good information Don.  I can get over 100 easily now. I have a big transmitter that will do 150+ with external audio gear for that kind of thing.  The T3 just runs with very minimal modifications.  This tube swap was sure worth the effort. I too never liked running 60 to 70% mod.

C

[Jim, W5JO]

A fellow out in Western Oklahoma has one and he had Peter Dahl wind him a custom modulation transformer to achieve the 100% level.  as I recall he made no other changes

[flintstone mop]

The Timtron mods and removing HV in the Tank circuit does the same thing. B'cast audio and a little over 100% pos. with the stock iron.
It will go higher if you load the PA. for 300 out. No one will notice the drop of 100 watts.

You can push it harder by using 4-400's as modulators. And that will require more cooling and a bigger filament xfrmr.
Fred...............I know coz I did it with a T368 about 10 yrs ago.

[ke7trp]

Fred is there any other reason to do that mod?  Safety?  I am pleased with the sound and do not play music here. I just use a D104.

The one area I am going to address on the T3 is the HV power supply. I noticed that the oil cans are leaking into the bottom deck.  I am going to remove the deck next time my brother is over to help me. I want to get the deck in the backyard. Then remove the caps and clean up the mess.

How/where do I dispose of those caps? 

A buddy has a 32ufd at 5000 volt can.  Thats probably severe over kill here, Maybe it will fit. I am sure Robert has something for me.  Maybe in the 15UFD range?  The T3 uses two 4UFD caps now after the choke for a total of 8ufd.

I spent an hour on the T3 today.  It sure modulates nicely now.  Such a simple swap and an enormous gain in audio.  I had it set right at 1475pep on the air with a 380 watt carrier. I used the T'3s audio limiter and set it to never go over 1500. Thats all I am allowed to run here and the T3 seems happy to do that now!



C

[KM1H]

You can push it harder by using 4-400's as modulators. And that will require more cooling and a bigger filament xfrmr.
Fred...............I know coz I did it with a T368 about 10 yrs ago.

The 4-400 is a 250 with fins, otherwise the same tube. You probably could get a bit more with PL-175A's

[flintstone mop]

You can push it harder by using 4-400's as modulators. And that will require more cooling and a bigger filament xfrmr.
Fred...............I know coz I did it with a T368 about 10 yrs ago.

The 4-400 is a 250 with fins, otherwise the same tube. You probably could get a bit more with PL-175A's

Good info thanks Carl.
Splains why Clark's fil. transformer runs hot.

[WD5JKO]

Splains why Clark's fil. transformer runs hot.

  I recall my Globe King 500 had two filament transformers in parallel for the 4-250 filament. Those transformers ran hot as hell, and I often had issues with low filament voltage to the RF Final tube. Some of this was socket contact resistance where the tube needed a nudge on occasion to light up. The other part was just low filament voltage @ 15 amps from them transformers. This would reduce off resonance peak current capability, and lower the modulation peaks. I ended up installing a larger 6V transformer with a variac on the primary. Problem solved.

   I got no idea about the T368, but for sure the extra filament draw of the 4-250's will lower the filament voltage. But how low? Not sure but if less than 4.75V, I'd not want to leave it that way for long. If the voltage is OK, maybe fuse that transformer primary in case it shorts out inside. Possibly a 2A standard fuse would do? The idea is to have protection since the rigs main fuse or breaker might be rated such that the filament transformer just burns up in case it faults.

Jim
WD5JKO

[The Slab Bacon]

The 4-400 is a 250 with fins, otherwise the same tube. You probably could get a bit more with PL-175A's

PL-175As are a kool tube. They WILL out perform 4-400s by quite a bit if all of the operating parameters are correct. However, they are getting to be unobtanium these days. That is also the reason my Invader 2000 will way out perform a T-bolt (both have what physically appears to be the same RF deck.) Just a few extra volts on the plate and screen make a VERY noticable difference in outpoot with the 175As. Without trying to plot the plate and grid curves, I wold guess that the 175As mst have a lot more emission.

[ke7trp]

My trans does not run hot jim. The T3 has a sturdy Filiment transformer.  It also has a filiment voltage meter on the front panel and a filiment adjustment knob.  Turning it straight on had a voltage Drop of .2 volts on the meter with the stock fil trans.  I turned the knob up to 5.0.  The manual states to run 5.0 to 5.2.  I have plenty of range left on the knob.  Tim said you do not need to change the transformer out but it would run hot.  If you add another transformer, there is no issue.

C

Splains why Clark's fil. transformer runs hot.

  I recall my Globe King 500 had two filament transformers in parallel for the 4-250 filament. Those transformers ran hot as hell, and I often had issues with low filament voltage to the RF Final tube. Some of this was socket contact resistance where the tube needed a nudge on occasion to light up. The other part was just low filament voltage @ 15 amps from them transformers. This would reduce off resonance peak current capability, and lower the modulation peaks. I ended up installing a larger 6V transformer with a variac on the primary. Problem solved.

   I got no idea about the T368, but for sure the extra filament draw of the 4-250's will lower the filament voltage. But how low? Not sure but if less than 4.75V, I'd not want to leave it that way for long. If the voltage is OK, maybe fuse that transformer primary in case it shorts out inside. Possibly a 2A standard fuse would do? The idea is to have protection since the rigs main fuse or breaker might be rated such that the filament transformer just burns up in case it faults.

Jim
WD5JKO

[k4kyv]

The Timtron mods and removing HV in the Tank circuit does the same thing. B'cast audio and a little over 100% pos. with the stock iron.
It will go higher if you load the PA. for 300 out. No one will notice the drop of 100 watts.

You can push it harder by using 4-400's as modulators. And that will require more cooling and a bigger filament xfrmr.
Fred...............I know coz I did it with a T368 about 10 yrs ago.

Fred is there any other reason to do that mod?  Safety?  I am pleased with the sound and do not play music here. I just use a D104.

From what I understand of the mod (don't have a schematic), taking the HV off the tank coil allows less rf by-pass capacitance plus loading capacitance to shunt the modulation transformer, thus producing a cleaner high end.  As I see it, one drawback to doing that is that the rf plate choke becomes more critical.  The reason for the original circuit is to avoid having an rf choke directly from the final amp plate to ground.  That is enough enough with a ham rig, avoiding resonances within a ham band, but the T-368 was designed for continuous frequency coverage from 1.5 mc/s up to something like 20 mc/s.  It would be virtually impossible to design a plate choke that did not have a resonant point somewhere in that range.  Tim's modification would work OK as long as the rig was used only on the amateur bands, but care must be taken in choosing the plate choke, just is it is with a typical all-band ham rig, to avoid resonances within one or more of the bands. With series feed, the choke can go to a low rf voltage point on the tank coil, where it is much less critical. Shunt feed relieves the modulator of most of the shunting effect of the tuning and loading capacitors, since the plate blocking capacitor can have relatively low capacitance and still be effective all way across the transmitter's tuning range.

I use series feed on both my homebrew rigs.  The choke goes from the mid-tap of the balanced tank coil, where there is theoretically zero rf, to the +HV line. My link-coupled final uses no loading capacitor to shunt additional capacitance across the mod transformer. Series feed requires the plate capacitor to be mounted on HV standoffs and use an insulated shaft coupling, since the full modulated +HV is on the frame of the capacitor.

It goes without saying that care must be taken in manually changing the final amplifier plug-in coils. I use both the Jesus Stick AND two layers of interlock protection with the HF-300 rig, and at least the Jesus Stick with the 8005 rig, which is not in a fully enclosed cabinet.

[ke7trp]

Thanks for the post Don.  I studied the schematic and I see that if you do the mod you will be modulating the plate of the tube directly.  In factory form, You are hooked to the other side of the coil at the low impedance tap I think.  I asked W0VMC if he thought it would help me in any way since I use a D104 mic, and he said he thought it would not be worth the effort.  The only real gain would be increased high end in the audio spectrum as you stated and at the cost of reduced frequency range.

C

[KM1H]

The Ameritron choke is 225uH and $20 and holds together well on RTTY or digi even on 160 at 1500W with a 1000pf bypass and 2000pf of blocking/coupling.

Some other designs with less L use the choke as a part of the 160M tank where its a very high impedance at some point down the choke.

Alpha recently went with a design similar to the Ameritron for their 2X 4CX1000A amp.
http://media.photobucket.com/image/alpha+8410+/wsmc551/Ham%20Radio/_PAS0127.jpg

[ke7trp]

I use that plate choke in the 2x4-400transmitter with great success.

[ke7trp]

Here is my power supply out and in the back yard.  Are the Transformers and choke filled with oil also?  The choke has oil all around it.  The HV insulators are dripping oil. The bottom of the Chassis is wet and has the largest amount of oil.  Its hard to see in the pics here. The top side oil has alot of dust and dirt on it (live in AZ).

I got a 12.5 UF 4000 volt cap comming.  I am going to use it to replace the two 4UFs.  The choke has me worried.  If its filled with oil like I think you guys are going to tell me it is, I will have to replace it also

C

[ke7trp]

Another picture.

[ke7trp]

The choke is fine! AlL the oil is from one of the 4 UF caps. The trans and the choke are leak free.  ALL cleaned up now and waiting for the cap to arrive!

C

[K3YA]

I bet it wall still play fine with only one 4uF capacitor in place if you feel the need to get on the air before the new one arrives.  Glad it wasn't the choke.

[ke7trp]

Thats how it was K3ya.  I just tested the caps,  One is 4.5 and other is.... 0.   Looks like i ran that way for years  

Gotta dyno tune a race car at 1 so no time to finish it today.  I will clean it up tomorrow, scrub the chassis down and then install the new 12.5/4kv cap

C

[WBear2GCR]

check the ripple with the bigger cap... I have had it happen where on some choke input supplies when the cap is made bigger, it does NOT help reduce ripple, it increases ripple. In which case you may need to "tune" the choke with a small cap in parallel.

Others may have more experience than I with this...

                   _-_-bear

[Ed/KB1HYS]

That tuned choke idea sound interesting. Ever try it? Supposed to make a high impedance at 60hz by resonating the choke with the parallel cap.  I seem to recall needing a fairly high voltage rating for the cap.

[WD5JKO]

Patrick Turner has a good analysis of power supply filter networks, including choke input:

http://www.turneraudio.com.au/powersupplies.htm

"Power supply resonance
Since we have an LC filter, it will have a series resonance and the RL of 800 ohms we have is not a low enough
R value to damp the resonance. For the power supply to have a flat response without a peak at Fo
RL must be = 1.41 ZC or ZL ( both equal reactance values at Fo ). This is something you need to remember for all LC
or CL low pass or high pass second order filters.
The LC filter is a low pass second order filter and behaves just like any other LC filter you may use at RF
or at AF in a speaker crossover. The frequency response of an under-damped LC filter will be flat at a couple of octaves below the Fo, and then a peak occurs at Fo, and then the response rapidly rolls off after Fo
at an ultimate rate of 12dB/octave."

The point is you want Fo well below the lowest modulation frequency, and certainly much lower than 120 HZ.

"Frequency of resonance for any LC circuit, Fo =  5,035  /   square root of ( L x C )     
                                                                                 
where Fo is in Hz,
5,035 is a constant for all equations to work,
C is in uF, and L is in milliHenrys." 

Jim
WD5JKO

[ke7trp]

The cap is after the choke. Should be ok.  Talked to a T3 owner that has a 20 UFD in there now. No trouble. Also common is putting in a larger cap in 610s. Same thing. No trouble, less hum.  I guess we will find out soon enough. Knowing my luck....   

ANyone got any tips for cleaning out that oil grime?   I spent an hour scrubbing with dish soap, then Fantastic.  Then hosed it out.   I got some out but not all. Its very very stuborn.  Dennis gave me some tips that helped but still alot of oil in and around every nook and cranny.

Tomorrow, I thought of trying Acetone and rags.  That might be my only option here.. Maybe some non clorinated Carb cleaner? 

The new cap is here.  ITs less then half the size of one of the factory 4 UF caps.  Same format.  I am going to cut the brackets down and put them in a metal brake tomorrow at work.  Then, the new 12.5 UF will mount like factory.    I have to extend two 14AWG wires and she is ready to be tested. 

Thanks for any tips.

[KM1H]

In a FW supply its 120 Hz that you want to filter and the cap better resonate it about 5% high unless you want fireworks as it comes up thru resonance at turn-on. Henry and a few others used resonant PS for years but as components aged failures became common as values changed.

Also its not easy to determine the actual Choke L with commonly available equipment or to select a precise resonating C value from whats available.