Balance modulator diodes

[w5rkl]

I have a question or two about what type of diodes are best for a balance modulator. I'm working on a Heathkit
SB-401 that has 2 of the 4 balance modulator diodes blown. I've check them with an ohm meter so I'm sure
they are bad, no high resistance regardless of the meter lead connecting to the diodes, all low resistance.

The diodes in questions have a color code of Brown White Brown. A website I found that has a diode color chart
and using that chart says these diodes are 1N191's. Mouser says a replacement for these diodes are the 1N4148
small signal fast switching diodes. I've found data information on the 4148's and some data sheets say nothing
about the construction (silicon verses germanium) and others say silicon. The good thing is the 4148 has higher
PIV rating than the 1N191. They both have the same forward voltage, 1 volt but the 4148's PIV is 100vdc
verses 90 volts for the 1N191. The current rating of the 4148 is also higher where as the 191 is 50ma.

I recall reading something a while back that germanium diodes are best for balance modulators verses silicon.
Any comments whether the 4148 will work in place of the 191's? I'd like to replace all 4 of them and if the
4148 should work, I'm in luck, I have quite a few of them here, all brand new.

Thanks for any help / comments anyone has.

73's
Mike
W5RKL

[WU2D]

Hi Mike,

It won't hurt to try the 1N4148's. I did the same thing in my Marauder which uses two and they worked great. If you want to get religious, try to match the forward voltage drop and select 4 that are close.

Mike WU2D

[W2JBL]

    i have used 1N4148 and 1N43 in my various HW101's and favorite SB102. 4148's work well but watch out- some (of both types) can't handle the level being applied to them, saturate, and the carrier balance drifts. the resulting signal is also distorted. my best results came from Radio Shack "genuine" 1N34's.

[WB3JOK]

You can get 1N270 (germanium diodes) 2 for $1.00 from allelectronics.com, .40 each in 10 or more quantity.

I'm not sure how important the increased forward voltage of the (silicon) 1N4148 is compared to the germanium. If there's a large enough carrier oscillator amplitude, and audio signal, it probably doesn't matter. But I'm no expert on what constitutes "large enough" 

It certainly does make a difference for the relative power output rectifier, and the diodes in the HM-102 wattmeter... "by actual experiment"  Anyway the 1N270 is a good replacement for a 1N191 and I'd use them. My two cents worth.

-Charles

[WA1GFZ]

1N270s are junk with poor leakage . A matched set of 1N4148s will work fine. 1N5711s even better.

[WB3JOK]

1N270s are junk with poor leakage . A matched set of 1N4148s will work fine. 1N5711s even better.

http://www.datasheetcatalog.com/datasheets_pdf/1/N/2/7/1N270.shtml

100 ua max reverse current at 50V, typical will be much less.  At the more usual 5-10V in a balanced modulator circuit would leakage be a major problem? Seems like the original circuit worked OK with germaniums... Anyhow if you say 1N4148's will work in that circuit and not affect the performance, there is certainly no reason to use the 270's...

-Charles

[WA1GFZ]

The 1n4148s have a bit higher junction voltage so make sure you have enough carrier level to make them switch. 1N5711 will work with the same drive as the 270 but much better performance.

[w5rkl]

Thanks to everyone for all the responses. 

I considered the 1N34's but it's PIV rating is alot lower than the 1N191 germanium diodes. The good thing is 1N34's are available everywhere for pennys but can they handle the voltage in the balance modulator? The SB-401 and the HW-101 have very similiar balance modulator circuits using very similiar components.

I also have quite a few diodes with color bands "Red Brown Black".  If the internet site is correct, that's a 1N210. The 1N210 has a  PIV rating, 45vdc, and a current rating, 27 mils, verses the 1N191 which has a 90 vdc PIV and 50 mil current rating.  The 1N4148 has a PIV rating of 100 vdc at 500ma current rating.

Although I have quite a number of 1N4148s, I'm concerned, since they are silicon, that the driving voltage may not be sufficient. I guess the only way is to try 4 of them to see. If the 1N210's are better then I have quite a few of them I could use. Finding a good data sheet on 1N210's is difficult. Most information I found has been bits and pieces, nothing like a regular data sheet. You can find companies selling them but I'd like to see a good data sheet on them before trying them. Plus I'm not sure the color code is actually a 1N210 since the internet site indicates inconsistancy in diode band colors. I've seen 1N191's with 2 BLACK bands so which is correct?

73's
Mike
W5RKL

[The Slab Bacon]

FWIW, I tried using 4148s in my Invader 2000, with no luck. I could not get the carrier to balence out properly, and also had strange looking distortion in the modulated waveform. (I also matched them for a uniform forward voltage drop)
 I also used the rat shak 1N34As and they worked ok-fine. i've also uesd 34A's in my KWM-2's and a few other SSB rigs with good results.

                                                        The Slab Bacon

[WA1GFZ]

Then you should look into 1N5711s because the lO drive is low.
These old diodes are pretty crappy performance compared to the stuff made in later years.

[Pete, WA2CWA]

As Frank said, 1N5711's are the way to go. These are Schottky type diodes and are designed for this type of operation. I replaced the balanced modulator diodes in my TS-830S with 1N-5711's and it made a world of difference in performance. You're wasting your time with 1N-4148's or similar types. The balanced modulator type circuitry in your rig was designed for germanium or the more recent Schottky type diodes. 1N5411's are available from a number of vendor sources and are not expensive.

Data Sheet
http://www.datasheetcatalog.com/datasheets_pdf/1/N/5/7/1N5711.shtml

[WA1GFZ]

There was a time when you could buy matched sets of 4 but I don't think it matters in this application.

[Pete, WA2CWA]

There was a time when you could buy matched sets of 4 but I don't think it matters in this application.

The price has come down so much for these things, you can buy a handful of 1N-5711's for cheap, and do some bench measurements, to find 4 that are close to similar specs.

[WB3JOK]

As Frank said, 1N5711's are the way to go. These are Schottky type diodes and are designed for this type of operation.

Looks good on the data sheet... far superior to an 1N270 indeed.

(Close your ears AMers )My HW-12, 22, 32 and 101, all use the same diode ring modulator circuit.
Is there a benefit to replacing the existing germanium diodes assuming they are currently working? Or is this a case of "if it ain't broke, don't fix it"?

-Charles

[Pete, WA2CWA]

At 10 cents each (Mouser web site) or 9 cents each (100 quantity), it's a good investment. If you get a reasonable quantity, you could do some mix and match in groups of four on the work bench.

[w5rkl]

I've decided to use the 1N5711's. It appears they are the best choice in this case. I may even change the diodes in my SB-400 as well. Again, thanks to everyone for all your help.

73's
Mike
W5RKL

[WA1GFZ]

I have 16 of them in a type 3A DBM in the first mixer of my homebrew RX driven by 1/2 watt of LO power. The diodes run just a bit warm......but the mixer kicks butt.

[K3ZS]

While on the subject.    Would the Schottky diodes be a good choice for a Foreward/Reflected (SWR) power meter?   I am always wiping out the 1N34's when I forget to put the meter in the high power position when running the amp.  Would the scale remain correct after recalibration if the newer diodes were used?   Would they be less prone to damage?

[WA1GFZ]

I have them in my power meter. Once a local lightning hit took out the reflected diode. 1N4148 will also work well unless you are monitoring QRP rigs.
I built a copy of the Collins watt meter with dual meters. FWD calibrated for 2 KW and reflected 200 watts. At  a quick glance I can tell if I'm under 2:1. (10% reflected)