[TenTec] Centurion 422b troubleshooting help needed

Mon Feb 22 03:43:22 EST 2016

Wow Gary, that was a lot of questions.

I generally use a 1:1 Maxwell choke in the two places I suggested.
Mine are home-brew but similar to the Model 8232 sold by THE WIREMAN.
http://thewireman.com/baluns.html#854 

Jim Brown k9YC has a paper on Coax Transmitting Chokes.  It's actually a ppt presentation.  
http://audiosystemsgroup.com/NCDXACoaxChokesPPT.pdf 

This serves a totally different purpose than the RF choke inside the amp that you were referring to.

I'm sorry but this is not the place to write all there is to know about what all common mode current can do inside a radio or inside the shack. It's simply too much to get into here.

But I will make one point:  the amp is not a faraday cage that blocks CMC from getting into it.  CMC can cause weird things to happen inside of any device.

I was not specifically speaking about the 2N3055 when I mentioned the NASA spec devices.  The 10% was just a rough number I picked.  Man that was 1967/68 and my grey cells are very faded grey there.  I can't remember specifics of any particular part.
It was a simple 2-dimensional X-Ray view.

What we looked for in the X-Ray was physical mis-configurations.
Say the wafer is supposed to be perfectly horizontal inside of the case.
Rarely was one perfectly horizontal.  Most were off by 2 to 20% and some were nearly 90% off - meaning a wafer that was supposed to be horizontal was almost vertical.  All the wires were attached and it measured good electrically, but it was way out of spec physically.  I cannot tell you what the spec was (what the limits to the angle were).  That was about 50 years ago.

When components like that are subject to mechanical shock and temperature swings, they are more likely to break down than those that were assembled properly.

Most likely quality got a lot better in later years, which would mean that the advice to find old components is not valid.  

I didn't stay at that job long; the Army decided they needed me more than NASA did.
There was nothing really highly qualified about doing my job.  It was something anyone could be taught to do - at least that part of it.  I also ran QC on the final products which required technical skills, but viewing component X-rays was not rocket science. 

My comment on 3055 was based on experience in the 70s and 80s when I was home-brewing a lot.  That was a very popular transistor, but as I said earlier, there were a lot of cheap no-names being sold and they popped like firecrackers.  It made sense to pay more for a name brand.

But as I said, a name brand is not a guarantee; it only assures you of a warranty.  It can still break prematurely. 

73 - Rick, DJ0IP
(Nr. Frankfurt, Germany)

-----Original Message-----
From: TenTec [mailto:tentec-bounces at contesting.com] On Behalf Of Gary J FollettDukes HiFi
Sent: Monday, February 22, 2016 8:47 AM
To: Discussion of Ten-Tec Equipment
Subject: Re: [TenTec] Centurion 422b troubleshooting help needed

I wonder what you mean by putting an RF choke on the output of an amp. 

There already is one, going directly from the center pin of the power output connector to ground. It is there to prevent presence of high voltage DC at the output connector should the plate blocking capacitor ever short. This is a safety device.

What type of RF choke do you use and specifically, where do you place it?

What is the “common mode current” pathway and how would this current affect lifetime of the keying transistor? Since the amplifier cabinet is a Faraday cage at HF frequencies, placement of external devices to suppress common mode currents would seemingly have no impact on devices contained within that Faraday cage.

I appreciate your sharing your experiences regarding the 2N3055 and your NASA experience. However, the Ten Tec amplifiers are a pretty small population. Statistically, even 10% fallout would be a pretty small number. There are likely several billion 2N3055 (and their cousins) transistors in service. This device has been in the marketplace successfully for over 40 years. They do fail from time to time but not in percentages even approaching 10%. Most cases of failure take place when they run hot for a long time (diffusion of materials within the transistor producing short circuit conditions) or when they are subjected to excess collector to emitter voltage (avalanche breakdown without current limiting).

Indeed, if anywhere near 10% of 2N3055 transistors failed in infancy, there would have been headline news about them, on a par with the electrolytic condenser plague of the late 1990’s.

In addition, the Ten Tec amp did not use the 2N3055, it used the MJ15001, which has a higher voltage rating (more than twice) and a higher current rating than the 2N3055, far more than one would expect to need for this menial application. 

In the absence of the issues I mentioned, a TIP 31C should suffice for this task at a lower cost.

>From a personal scientific interest perspective, since I use high resolution 3D industrial Xray tomography from time to time for defect analysis, I wondered what you looked for in the X-rays of 2N3055 transistors to give you the go or no go acceptance criterion? What defect could you see via the 2 dimensional Xray methods available then that would predict device performance or lifetime?

Thank you for sharing,

Gary

W0DVN

> On Feb 22, 2016, at 1:04 AM, rick at dj0ip.de <Rick at DJ0IP.de> wrote:
> 
> The point of looking for the cause is well taken, but you guys may be thinking in the wrong direction.
> 
> There is a possibility that common mode current is the cause of the breakdown.
> To be on the safe side, I would place a good RF Choke directly on the output of the amp.  I do that all the time.  I also use one between the amp and the transceiver.
> 
> 73 - Rick, DJ0IP
> (Nr. Frankfurt, Germany)
> 
> 
> 
> -----Original Message-----
> From: TenTec [mailto:tentec-bounces at contesting.com] On Behalf Of Gary J FollettDukes HiFi
> Sent: Monday, February 22, 2016 5:25 AM
> To: Discussion of Ten-Tec Equipment
> Subject: Re: [TenTec] Centurion 422b troubleshooting help needed
> 
> I can only imagine that using legacy old inventory would be the best choice if indeed manufacturing issues are the root cause. These are readily available on the “world’s largest auction site”.
> 
> However, with a transistor with 20 amp capability is keying a line with a maximum of 1 amp of current, and there appear to be multiple examples of failures, I have to wonder where the problem lies.
> 
> One could speculate that using a transistor  in this capacity is a good idea at all. The reason is that, just as was the case with Cathode keying of old tube transmitters, the Cathode of a power tube in an amplifier can float to voltages that are pretty high, and this is what caused key clicks in those old Cathode keyed transmitters. If, by metaphor, the collector of the transistor in this keying scenario were to float up above the breakdown voltage of the transistor, then the transistor would act sort of like a Zener diode and pass current with power dissipation equaling the product of the Zener potential times the measured residual current flow, whatever that number might be. That transistor could run fairly warm…
> 
> It might be interesting to measure the voltage that appears at the collector during key up and determine if this is taking place. In addition, a current measurement of the current through the transistor (in a properly working amplifier) would be helpful.
> 
> I know a fix for the problem if it exists but won’t waste time with it until a need arises.
> 
> Gary
> 
> W0DVN
> 
> 
> 
>> On Feb 21, 2016, at 10:05 PM, A R <raf_3 at msn.com> wrote:
>> 
>> 
>> Gary,
>> 
>> No problem here. I have a built-in autocorrect "uncorrector". I read "overcurrent" despite the autocorrect's corruption.
>> 
>> I'm inclined to suspect the failures are due to a device mfg quality issue...like doping impurities, and probably lot-related. Wish I had the date codes from the failed devices (assuming that they were all MJ15001's). The one in my amp carries the "ON" logo, is from Mexico, and has a lot code of "BM0318". I believe that Harris also makes the MJ15001. Too bad Motorola is no longer mfg such devices here in the valley. Do we have any on-shore manufacturers of semi's anymore?
>> 
>> 73,
>> 
>> Allen--W7GIF
>> 
>> 
>> ----- Original Message ----- 
>> 
>> From: Gary J FollettDukes HiFi<mailto:dukeshifi at comcast.net> 
>> 
>> To: Discussion of Ten-Tec Equipment<mailto:tentec at contesting.com> 
>> 
>> Sent: Sunday, February 21, 2016 8:25 PM
>> 
>> Subject: Re: [TenTec] Centurion 422b troubleshooting help needed
>> 
>> 
>> 
>> Darned AUTOCORRECT!
>> 
>> “overcurrent”, NOT “overcorrect”!
>> 
>> 
>> 
>>> On Feb 21, 2016, at 9:17 PM, Gary J FollettDukes HiFi <dukeshifi at comcast.net<mailto:dukeshifi at comcast.net>> wrote:
>>> 
>>> This transistor sounds like a likely candidate as the amp is acting like it is keyed regardless of the state of the keying line.
>>> 
>>> Question is, why did it fail? A 2N3055 is good for 15 to 20amps so overcorrect is not a factor. Is there adequate transient suppression around this transistor to protect it from voltage transients?
>>> 
>>> Just curious…
>>> 
>>> Gary
>>> 
>>> W0DVV
>>> 
>>> 
>>>> On Feb 21, 2016, at 8:03 PM, Carter <k8vt at ameritech.net<mailto:k8vt at ameritech.net> <mailto:k8vt at ameritech.net<mailto:k8vt at ameritech.net>>> wrote:
>>>> 
>>>> On 2/21/2016 7:42 PM, A R wrote:
>>>>> My Centurion 422b amp is sick. With initial application of power, in selected PTT and STANDBY modes (and un-keyed), the 3-500's are drawing 140 ma plate current (and barely perceptible grid current). Panel MODE led's indicate proper mode when switched between STANDBY and OPERATE, and the TRANSMIT led is unlit. When placed in OPERATE mode, and keyed (at the PTT RCA jack), the TRANSMIT led is illuminated. Keying the PTT input when in STANDBY mode does NOT illuminate the TRANSMIT led. So, the panel MODE-indicating led's suggest that mode control functions are OK. The RF input/output switching/routing is correct for all modes.
>>>>> 
>>>>> 
>>>>> Have begun troubleshooting the QSK board, and Filament-AC board (with the tube bias circuitry), and the HV filter board, but have found that the componentry on those boards differs from that described in the 422 (no-suffix) manual. Although the manual available on the RKR website is identified as a 422b manual, it appears to be for a 422 (no-suffix) amp. In that manual, there are discrepancies between the schematic and board layout and circuit trace views for the FILAMENT-AC, QSK, and HV FILTER boards. And, the QSK board in this amp is a p/n 78247, not the 81542 called out in the 422 manual.
>>>>> 
>>>>> 
>>>> 
>>>> Sure sounds like the bias transistor on the Filament-AC board. Been there, done that.
>>>> 
>>>> Let us know how you make out...
>>>> 
>>>> 73,
>>>> Carter   K8VT
>>>> _______________________________________________
>>>> TenTec mailing list
>>>> TenTec at contesting.com<mailto:TenTec at contesting.com> <mailto:TenTec at contesting.com<mailto:TenTec at contesting.com>> <mailto:TenTec at contesting.com<mailto:TenTec at contesting.com> <mailto:TenTec at contesting.com<mailto:TenTec at contesting.com>>>
>>>> http://lists.contesting.com/mailman/listinfo/tentec<http://lists.contesting.com/mailman/listinfo/tentec> <http://lists.contesting.com/mailman/listinfo/tentec<http://lists.contesting.com/mailman/listinfo/tentec>> <http://lists.contesting.com/mailman/listinfo/tentec<http://lists.contesting.com/mailman/listinfo/tentec> <http://lists.contesting.com/mailman/listinfo/tentec<http://lists.contesting.com/mailman/listinfo/tentec>>>
>>> _______________________________________________
>>> TenTec mailing list
>>> TenTec at contesting.com<mailto:TenTec at contesting.com> <mailto:TenTec at contesting.com<mailto:TenTec at contesting.com>>
>>> http://lists.contesting.com/mailman/listinfo/tentec<http://lists.contesting.com/mailman/listinfo/tentec> <http://lists.contesting.com/mailman/listinfo/tentec<http://lists.contesting.com/mailman/listinfo/tentec>>
>> _______________________________________________
>> TenTec mailing list
>> TenTec at contesting.com<mailto:TenTec at contesting.com>
>> http://lists.contesting.com/mailman/listinfo/tentec<http://lists.contesting.com/mailman/listinfo/tentec>
>> _______________________________________________
>> TenTec mailing list
>> TenTec at contesting.com
>> http://lists.contesting.com/mailman/listinfo/tentec
> 
> _______________________________________________
> TenTec mailing list
> TenTec at contesting.com
> http://lists.contesting.com/mailman/listinfo/tentec
> 
> _______________________________________________
> TenTec mailing list
> TenTec at contesting.com
> http://lists.contesting.com/mailman/listinfo/tentec

_______________________________________________
TenTec mailing list
TenTec at contesting.com
http://lists.contesting.com/mailman/listinfo/tentec