Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
More robust MPN3404 PIN diode for TR7?




 
The AM Forum
October 22, 2017, 09:38:26 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: More robust MPN3404 PIN diode for TR7?  (Read 7392 times)
0 Members and 1 Guest are viewing this topic.
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3189


Rick & "Roosevelt"


« on: May 10, 2008, 12:49:29 PM »

Forgive all the posts and answers today, must be Saturday Grin

Anyway I just popped another MPN3404 PIN diode in the TR7 Assembly 1400 high pass filter module.  They're readily available from Mouser but I was wondering if there's a more modern, beefier equivalent? 

A certain one, CR 1402 seems to blow about every 5 to 10 years or so since I've owned the rig, last time in '98.

 Must be a farily common failure; first time (1980) I called Bill Gaible  of Drake and he sent me two or three diodes along with location diagrams when I simply described low output on 10, 15 and 20 or so.
Logged

RICK  *W3RSW*
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #1 on: May 11, 2008, 08:30:23 PM »

Gee I've owned a TR7 since early 80s and never lost one. Do they open up?
How about 2 in parallel to carry more RF current. BTW a 1N4007 works pretty well as a pin above 5 MHz or so depending on Trr. 
Logged
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3189


Rick & "Roosevelt"


« Reply #2 on: May 12, 2008, 09:55:12 AM »

yeah, I could try a 1N4007.  PIN's are supposed to have very low capacitance. Anyway, perusing the catalogs, seems like a lot of PIN's with greater current ratings, higher PIV, lower capacitance, etc. so I was wondering if anyone had tried something different than the stock ones?

Is yours a TR7A?  later production run than my #7130 TR 7?
What makes me suspicious is that the Drake guy in 1980 sent, gratis, 3 each of two different types of PIN's depending on where they go in the circuit along with marked up zerox copies of where they go, the  in and out of circuit resistance measurements, etc.
He was either a very nice guy or some of the radios have a problem.

-Like he kept them under the counter for twice daily calls or something..   

Last time something like this happened, the ignition switch in my JD mower failed, I took it into the parts dealer and he pulls one out from under the counter.  (made in Mexico, but has lasted longer than original.)  .. No looking up part no's on the computer or anything. - Has to be a very high failure rate item or maybe common to a lot of equipment or both.
Logged

RICK  *W3RSW*
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #3 on: May 12, 2008, 11:24:39 AM »

My unit is a low serial number under 500. It is a TR7A. I wonder if your band switch is not lined up properly or the bias is wrong to the diode. Check MaCom for different diodes. Just pick a diode with the same or longer Trr and same or greater wattage rating. Also make sure the breakdown voltage is similar. If it remember these Pins are upstream of the final module where the RF levels are pretty low. A couple years ago I pulled my unit apart and sprayed deoxit on all board connector contacts. I lost RX and tX power was unstable. Last time i used it things were fine. fc
Logged
Rob K2CU
Contributing
Member
*
Offline Offline

Posts: 346


« Reply #4 on: May 12, 2008, 12:04:03 PM »

CR1402 is one of two series connected PIN diodes, the other CR1401, that provide a connection of the calibrator signal to the internal side of the high pass filter module. In normal operation, this diode is not biased on. Current rating is not an issue here as the bias current when the calibrator is turned on is less than 5 mA. There is no RF power running through the diode, only calibrator signal. IF the diode fails shorted, then a .001 uF capacitor that is between the junction of the two diodes and ground will load down RF in both receive and transmit. The reason for two diodes and the capacitor instead of just one diode is unknown. The calibrator divider IC is only enabled in calibrate mode, and its output connected to CR1401 through a 100 Ohm resistor.

CR1402 is exposed to the outside world as it is at the junction where the VLF input comes in. IS it possible that ESD gets to the part? One must be very diligent in handling these parts and causing latent Electrostatic Damage to them that will show up months or even years later.

Frank is quite right that you could use a 1N4007 diode. But, if you are in a pinch to replace the part, consider this: If you have the digital display, the calibrator is irrelevant; Remove CR1402 and leave it out for now. OR, rempve both CR1402 and CR1401 and put the good part that was CR1401 in the CR1402 position and a 1N4007 in the CR1401 position.

Logged
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3189


Rick & "Roosevelt"


« Reply #5 on: May 12, 2008, 12:09:59 PM »

Very informative and shows that RF thiefing is the reason for lower output. Yes, bad diode measured 8 ohms, both directions in circuit and good diode shows the expected 70 ohms/infinity.
Thanks.
Logged

RICK  *W3RSW*
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #6 on: May 12, 2008, 12:29:32 PM »

Gee,
Might be a good idea to hang a 10K resistor across the VLF input to bleed off any static charge. I would think the junction fairly large on a HF pin so must take a bit of a jolt to hurt it. Rob 5 volts on 100 ohms is 50 ma. Maybe the calibrator is nailing the diode. I think changing the 100 ohm calibrator isolation to 1000 ohms might be a good idea. It could even be a turn on transient when the radio power up. Since the Rf level is low I suspect it a DC problem killing the diode. Can the cap hold a DC charge and turn off dump itself through the diode??
Logged
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3189


Rick & "Roosevelt"


« Reply #7 on: May 12, 2008, 01:50:58 PM »

Good thought on static drain on VLF ant input which actually comes from a combo cinch jones connector (accessory socket)adjacent to a bunch of other stuff including the PTT line, VOX enable, etc.  With a lot of transmission from 300+ watt AM rig into a not very far away windom, who knows what garbage gets on that pin..?  On second thought it's wired through a .05u cap and 1k r into our friendly diode's junction, so I'll just ground the VLF pin.

I also thought about a 5pf cap across the diode to grab fast transients. (lot less drain than a .001uf at these low impedances.)

So until the Mouser order gets here, I'll disconnect CR 1402.

Thanks both; you've clarified my thinking.
Logged

RICK  *W3RSW*
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #8 on: May 12, 2008, 02:10:55 PM »

Reverse voltage ratings range from 15 to 20 volts depending on source. I would think transients slowly turn it into a resistor. My rig wants a good ground or the TX is unhappy.
Logged
Rob K2CU
Contributing
Member
*
Offline Offline

Posts: 346


« Reply #9 on: May 12, 2008, 03:35:45 PM »

Frank, the DC bias current for the two diodes comes from the +10VDC calibrate enable signal working through both diodes and two 1 K resistors, for a value of less than 5 mA DC. The 74LS390 four bit counter that provides the 25 KHz square wave is actually working through the 100 Ohmer in series with a .05 uF cap and one of the two afore mentioned 1 K resistors. EWB simulation suggests a minimum current of about 3 mA and a max of 6 mA from the bias plus the square wave when the calibrator is on. It is this modulation of the diode current in the very nonlinear portion of the diode curve that produces all those birds and spurs if you mistakenly leave the calibrator on. The VLF input has a series cap with, believe it or not, only 16V rating! it also has a 1K resistor in series, part of the famous 20 dB pad for VLF. The diode is certainly in danger from static on the VLF input. But, unless the subject radio had that input connected to, or had been worked on with power applied and had boards or connectors plugged/unplugged while powered, etc, I still suspect latent ESD from component handling.

I just had yet another display segment die over the weekend, so now I have to design a replacement display for the radio. I have found correctly sized red or blue devices for the project. any preference? the .3 inch digit size is the tough part.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #10 on: May 12, 2008, 05:10:52 PM »

Rob,
I don't have the schematic in front of me but was thinking of the digital drive through the 100 ohmer. LS source current is low so forget what I said.
Bummer on the displays. Check Jameco they had a selection of displays back a while ago. Are they stand alone or built in decoders?
Logged
John K5PRO
Member

Offline Offline

Posts: 1026



« Reply #11 on: May 12, 2008, 11:47:20 PM »

Check this guy overseas for replacement displays

http://www.df4nw.de/44.html
Logged
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3189


Rick & "Roosevelt"


« Reply #12 on: May 13, 2008, 10:05:52 AM »

Way cool John; I like the two color combo. 
He has some pretty nice pre-amps too.

BTW I lifted CR 1402 ; sensitivity and output came back better than before the diode totally crapped out.  Might even be better than new.  Now I'm a believer about the degradation over time, improper static handling, latent damage theory of these things.  I think Drake knew of the potential problems of amateur repairs and installation and sent me a whole bunch just in case.  Been too long ago but I might have even been cautioned about installation static in the phone conversation w/DRake.

I may have degraded the last replacement diode years ago simply by ohm-meter checking to get base line meas. of a new diode.  Makes one wonder about using a modern digital Fluke or equiv. in a simple auto ohms, auto ranging check as far as initial voltage across the diode.

 At any rate Rob, your right on in sensitivity to handling. Next diode to go in will not be "pre-tested" or fooled with anymore than simple replacement by grounded wrist strap method.
Logged

RICK  *W3RSW*
Rob K2CU
Contributing
Member
*
Offline Offline

Posts: 346


« Reply #13 on: May 13, 2008, 01:14:09 PM »

Rick,

You could try wrapping a piece of fine wire around the leads to short them until you have installed the diode in the radio. then pull off the wire. some transistors from long ago in my stash have fine spring wire around the four leads with a small loop at end to easily pull off after installation.

Thanks for links to displays, will check them out. The problem has been fitting into the small cutout in front panel.
Logged
John K5PRO
Member

Offline Offline

Posts: 1026



« Reply #14 on: May 13, 2008, 04:21:09 PM »

Have any of you done the AM mod to TR7 that was published in Electric Radio last summer? I have all the parts for it (the filter) but still need to finish troubleshooting a dead transmit side in mine before I make the changes. It looks like the TR7 with the mod may be a nice exciter for AM if you drive a bigger power amplifier with it.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #15 on: May 13, 2008, 04:31:05 PM »

MRF421s work well in the final of the TR7. I think i put MRf 475s in the driver.
My unit was killed by a CBer idiot before I got it and the finals and drivers were blown. fc
Logged
WA7DUY
Member

Offline Offline

Posts: 3


« Reply #16 on: September 23, 2017, 01:54:38 PM »

Might be a good idea to hang a 10K resistor across the VLF input to bleed off any static charge.
I thought that was the job of the gas discharge tube (GDT) across the antenna connection on the back of the TR7.  I see Drake used a much lower firing voltage for the GDT than Kenwood is using now (50v compared to 400v).
Logged
KA2DZT
Member

Offline Offline

Posts: 1967


« Reply #17 on: September 24, 2017, 08:38:40 AM »

My club bought a TR7 when they first came out.  Rig worked fine for a number of years.  Developed a problem with very low output on 10,15 and 20.  Everyone thought it was the finals and wanted to have them replaced.  I disagreed and took the rig back to my shop.  After a few days of trouble shooting I determined that there was low drive on those bands.  Trace the problem to shorted diodes in the crystal calibrator circuit.  Fixed the problem with a couple of 3 cent 914 diodes.

Fred
Logged
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3189


Rick & "Roosevelt"


« Reply #18 on: September 24, 2017, 02:06:28 PM »

Amazing retro.
You never know when a subject will be useful again.
Glad you researched before replacing major trace lifting transistors.  I replaced them once unnecessarily too when Drake was still in business so did lift a trace from soldering heat removal.

Before getting a K3 and then some SDR gear, I'd left the TR7 pretty much always on casual receive for many years, since '76 I think, and became quite fond of it, barring a few little easily reparable quirks like the diode problem and replacing or beefing up the power supply caps. Oh, yeah and replaced the quad diode smaller supply rectifiers in same.

Used it so much the blue faded to yellow on the display. Bought the replacement sheet but never got around to changeout.  I fire it up now and then for SWR checks of new wire antennas at low power.  Yeah, did the freq. Logic bypasses and cuts, um hmm.  Grin

Freq. readout always a hair off after warm up. Wanted to put in the digi freq. synchro. locking circuits found on the web and probably still available but didn't even see the need until exact freqs.  expected became the norm. No one cried much then, just used xit and rit, and ate the diff. If not too objectionable in roundtables.  In AM of course, no one cared.
--Then along came syncro AM, whistle stop crying from newer hams or perfectionists.  Grin

Long live the TR-7 !
Logged

RICK  *W3RSW*
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 6704



WWW
« Reply #19 on: September 25, 2017, 12:09:18 AM »

synchro or not does not bother me. Sometimes I use the Sherwood, sometimes not and set the bandwidth wide enough- all standard AM receivers and a couple plastic radios that still have the 'rectifier' type detector. no problems!
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.056 seconds with 18 queries.