ICOM756pro2 bias compensation diode - UK Vintage Radio Repair and Restoration Discussion Forum

Colinaps · 4th Mar 2023, 5:55 pm

Afternoon all,

I'm just replacing a pair of properly popped 2SC1972 driver transistors in a 756pro2.
The thermal compensation diode is O/C, at least according to all my DMMs.

The manual says the original is an MA29B, shown as a diode on the schematic, which makes perfect sense but any search for info takes me down a rabbit-hole of esoteric (to me!) thermistors. I feel something like Farnell part 3780842 should work, but none have I, and I'm not keen on making up an order for one microscopic component.

The original looks a lot like a surface-mount 1N4148 on a tiddly bit of FR that was submerged in a pile of heatsink paste, but the middle of the glass body sits on a little plinth of solder. Many other radios use a bog-standard silicon diode as D1 here, to back off the bias on the pre-driver as the driver gets hot. Is there any reason why it shouldn't be replaced with an everyday silicon diode, or maybe two legs of a small metal transistor?

Cheers,

Colin mm1aps

Colinaps · 4th Mar 2023, 6:20 pm

Or perhaps...
It actually has a forward resistance of just under 10M at room temp, which drops when a soldering iron approaches.
Seems a bit high considering it has R6 (220R) in parallel. Perhaps it only comes into play just before the smoke... after, in this instance.

Colin.

frsimen · 4th Mar 2023, 7:03 pm

Data for the MA29 series of components is a bit confusing. It seems to be a variety of diode but is described as a variable resistor.

Some in the MA29 range comprise several diodes in series, so will have a higher forward voltage than you would expect. What it isn't is a thermistor and the Farnell part you have mentioned is not suitable. I don't know what it can be replaced with.

Paula

Cruisin Marine · 4th Mar 2023, 7:11 pm

https://html.alldatasheet.com/html-p...03/1/MA29.html

the B seems to be an ordinary 0.6v diode according to this

Radio Wrangler · 4th Mar 2023, 7:21 pm

Something is wrong with that diagram. The second stage, the push pull pair, has NO bias applied to it.

What looks like it should be the bias supply is shown as feeding the collector of the first stage, but that collector already has a beefy power feed from a supply rail via L2. So this power feed is shorted to ground by D1, which wouldn't survive the experience

I think someone's made a muck up of the drawing of the interstage transformer. It would be ever so slightly more sensible with the centre-tapped winding feeding the bases of the second stage and the solitary winding handling the collector of the first stage.

As the driver pair have both blown, it's likely that collector power came backwards out of their base connections, and your diode tried to clamp it, burning out in the process.

The part you need really is a silicon diode and not a thermistor. It doesn't need to be fast or low leakage or anything special, but it does need to have a reasonably similar area to the original, so that the voltage-current curve at DC is similar e bias twiddle pow will get you some adjustment, but you may need to try a couple of different current ratings of diode and see what tracks best over temperature.

The quick way out is to order one from ICOM UK.

David

Colinaps · 4th Mar 2023, 8:21 pm

Thanks everyone,

Definitely something wrong with the drawing - adjusting R11 slightly changes the resistance from the drivers' bases to ground with link W29 (13.8V driver supply) disconnected. I

I did what should have been obvious if I'd been paying attention and lifted one of the same type of diodes from the PA, and right enough, its forward drop was 0.6V, about 500k resistance at room temp.

So it'll be getting an ordinary diode for now then and we'll see what the power does as it heats up. I suppose two in series would shut it down at a lower temperature, once it's been set up at room temp?

Cheers,

Colin.

Radio Wrangler · 4th Mar 2023, 10:17 pm

two in series will overbias the stage, which will run silly current and you're back to buying more transistors.

These bias diodes run in a SHUNT position, compensating theVbe of the RF devices. Think of them and the RF devices making a sort of rough current mirror. The current through the bias trimmer goes through the diode and establishes a voltage. That voltage biases the base-emitter junction of the RF devices a little bit forwards. The idea is that the b-e junction and the diode junction are at the same temperature and therefore the transistor bias current will track the diode current over temperature.

Add a second diode, the transistor sees doubled bias voltage and its current is relatively huge. Next comes the smoke if nothing fails faster

David

Colinaps · 5th Mar 2023, 1:27 am

Ah - somewhere in my little brain, R11 was going to earth - but it isn't, it's 8v on tx. I find it almost impossible to follow a circuit or any sort of plan on a computer screen. Is it just me?

The finals would seem to have a sensible scheme, with a linear power supply to provide the bias, and a sensing diode on the pass transistor in series with that on the final. Of course if the pass tranny (Q8) fails s/c, that's the end of the finals....

I shall try it out in daylight, probably in a couple of days when my wife's away at work and can't smell the smoke....

Cheers,

Colin.

4th Mar 2023, 5:55 pm	#1
Colinaps Pentode Join Date: Jul 2009 Location: Roxburghshire, UK. Posts: 196	ICOM756pro2 bias compensation diode Afternoon all, I'm just replacing a pair of properly popped 2SC1972 driver transistors in a 756pro2. The thermal compensation diode is O/C, at least according to all my DMMs. The manual says the original is an MA29B, shown as a diode on the schematic, which makes perfect sense but any search for info takes me down a rabbit-hole of esoteric (to me!) thermistors. I feel something like Farnell part 3780842 should work, but none have I, and I'm not keen on making up an order for one microscopic component. The original looks a lot like a surface-mount 1N4148 on a tiddly bit of FR that was submerged in a pile of heatsink paste, but the middle of the glass body sits on a little plinth of solder. Many other radios use a bog-standard silicon diode as D1 here, to back off the bias on the pre-driver as the driver gets hot. Is there any reason why it shouldn't be replaced with an everyday silicon diode, or maybe two legs of a small metal transistor? Cheers, Colin mm1aps Attached Thumbnails

4th Mar 2023, 6:20 pm	#2
Colinaps Pentode Join Date: Jul 2009 Location: Roxburghshire, UK. Posts: 196	Re: ICOM756pro2 bias compensation diode Or perhaps... It actually has a forward resistance of just under 10M at room temp, which drops when a soldering iron approaches. Seems a bit high considering it has R6 (220R) in parallel. Perhaps it only comes into play just before the smoke... after, in this instance. Colin.

4th Mar 2023, 7:03 pm	#3
frsimen Heptode Join Date: Jan 2011 Location: Croydon, London, UK. Posts: 773	Re: ICOM756pro2 bias compensation diode Data for the MA29 series of components is a bit confusing. It seems to be a variety of diode but is described as a variable resistor. Some in the MA29 range comprise several diodes in series, so will have a higher forward voltage than you would expect. What it isn't is a thermistor and the Farnell part you have mentioned is not suitable. I don't know what it can be replaced with. Paula

4th Mar 2023, 7:11 pm	#4
Cruisin Marine Octode Join Date: Jul 2018 Location: Worthing, West Sussex, UK. Posts: 1,003	Re: ICOM756pro2 bias compensation diode https://html.alldatasheet.com/html-p...03/1/MA29.html the B seems to be an ordinary 0.6v diode according to this __________________ "Behind every crowd, there's a silver Moonshine"

4th Mar 2023, 7:21 pm	#5
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,901	Re: ICOM756pro2 bias compensation diode Something is wrong with that diagram. The second stage, the push pull pair, has NO bias applied to it. What looks like it should be the bias supply is shown as feeding the collector of the first stage, but that collector already has a beefy power feed from a supply rail via L2. So this power feed is shorted to ground by D1, which wouldn't survive the experience I think someone's made a muck up of the drawing of the interstage transformer. It would be ever so slightly more sensible with the centre-tapped winding feeding the bases of the second stage and the solitary winding handling the collector of the first stage. As the driver pair have both blown, it's likely that collector power came backwards out of their base connections, and your diode tried to clamp it, burning out in the process. The part you need really is a silicon diode and not a thermistor. It doesn't need to be fast or low leakage or anything special, but it does need to have a reasonably similar area to the original, so that the voltage-current curve at DC is similar e bias twiddle pow will get you some adjustment, but you may need to try a couple of different current ratings of diode and see what tracks best over temperature. The quick way out is to order one from ICOM UK. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

4th Mar 2023, 8:21 pm	#6
Colinaps Pentode Join Date: Jul 2009 Location: Roxburghshire, UK. Posts: 196	Re: ICOM756pro2 bias compensation diode Thanks everyone, Definitely something wrong with the drawing - adjusting R11 slightly changes the resistance from the drivers' bases to ground with link W29 (13.8V driver supply) disconnected. I I did what should have been obvious if I'd been paying attention and lifted one of the same type of diodes from the PA, and right enough, its forward drop was 0.6V, about 500k resistance at room temp. So it'll be getting an ordinary diode for now then and we'll see what the power does as it heats up. I suppose two in series would shut it down at a lower temperature, once it's been set up at room temp? Cheers, Colin.

4th Mar 2023, 10:17 pm	#7
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,901	Re: ICOM756pro2 bias compensation diode two in series will overbias the stage, which will run silly current and you're back to buying more transistors. These bias diodes run in a SHUNT position, compensating theVbe of the RF devices. Think of them and the RF devices making a sort of rough current mirror. The current through the bias trimmer goes through the diode and establishes a voltage. That voltage biases the base-emitter junction of the RF devices a little bit forwards. The idea is that the b-e junction and the diode junction are at the same temperature and therefore the transistor bias current will track the diode current over temperature. Add a second diode, the transistor sees doubled bias voltage and its current is relatively huge. Next comes the smoke if nothing fails faster David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

5th Mar 2023, 1:27 am	#8
Colinaps Pentode Join Date: Jul 2009 Location: Roxburghshire, UK. Posts: 196	Re: ICOM756pro2 bias compensation diode Ah - somewhere in my little brain, R11 was going to earth - but it isn't, it's 8v on tx. I find it almost impossible to follow a circuit or any sort of plan on a computer screen. Is it just me? The finals would seem to have a sensible scheme, with a linear power supply to provide the bias, and a sensing diode on the pass transistor in series with that on the final. Of course if the pass tranny (Q8) fails s/c, that's the end of the finals.... I shall try it out in daylight, probably in a couple of days when my wife's away at work and can't smell the smoke.... Cheers, Colin.