SHP2 Amp

[Lincoln]

I recapped an SHP2 that had no sound when I started. I sounds good now, but the heat sinks are getting hot. I tried adjusting the bias and it was reading around 1.805 volts. I couldn't get the reading to change using the potentiometers, so I put new ones in. I used NTE 100 ohm pots. I checked before I put them in and could measure 0 to 100 ohms. With the new pots, I still can't get the bias reading to change. I also found that with the wires hooked up to the pots they would only read 35 ohms. With power to the amp the resistance dropped even further. Any ideas where I should start looking for the problem?

[robnyc]

The problem is likely under the affected channel's heatsink. On the 2's and 3's they use a pair of diodes and a thermistor along with a low-ohm resistor inside there.

The odd thing in your post is the 1.805 reading. I set them to 1.5 -2mv so that should be fine. They idle stone cold. Make sure you are reading the bad channel.

RobNYC

[Ron Rich]

Or leaky drivers--Don't know about the -2, but all but the -3, CODE C, originally had the transistor based bias system. Ron Rich

[Lincoln]

Do amps sometimes take a while to "settle in" after recapping? I came home today, played one song, both sides of the heatsink got hot. Went and mowed my lawn. Came back, checked that bias reading and was able to adjust it down, meaning the bias pots responded like they should. Now I just played several songs, and no heat on the heatsink.

Ron do you mean leaky driver transistors?

[robnyc]

May 16, 2018 18:56:24 GMT Ron Rich said:

Or leaky drivers--Don't know about the -2, but all but the -3, CODE C, originally had the transistor based bias system. Ron Rich

Ron, here is the only print I have on the SHP-2:
www.fileconvoy.com/dfl.php?id=g91b312ed95230d2e1000083124683160e00fdfb1c3  

It shows the same bias ckt as the 3. Another point is that the connection for the SOS sampling point at the output emitter (P5105 -15)is "N.C."    I wonder if they also eliminated the associated parts on the boards as well.

If so the "logic" must have been that the higher current handling of the finals against the weaker power supply would serve to limit the potential for catastrophic failure. This sort of thing was common in consumer stuff back then as well.

RobNYC

[Ron Rich]

Hi Rob,
I can not access that file, and do not have a book on that amp, handy--but I know there was no "overload" ("SOS") in the -2's--Ron Rich

[robnyc]

May 17, 2018 3:02:57 GMT Ron Rich said:

Hi Rob,
I can not access that file, and do not have a book on that amp, handy--but I know there was no "overload" ("SOS") in the -2's--Ron Rich

It does appear that the SOS parts are on the preamp board but I don't seem to have the driver bd.
Here is a jpg of the finals:

Rob

Attachments:

[benp]

Hi all,

From my experience with the shp amps you need to clean the bias controls and make absolutely sure they work correctly before doing anything else. If it played before the recap it will play after as well. For safety reasons if you use a dim bulb and it burns brightly unplug the amp and jump out the bias pots then try it again. Jumping out the bias pots will not hurt the circuit but the amp will sound a bit distorted and run cold telling you the bias pots are set too high, dirty, or open.

There are three different schematics for the shp amps and they can apply to any of the three shp1, shp2, shp3. I have found this to be true after burning up a bunch of them over the course of my career. The boards can't just be swapped between amps because some used a transistor under the heat sync in some models and a thermister that looks like a red disc capacitor on others and the boards have different components depending on the type of heat sensing components they used. There are other component differences between models as well.

Good luck,

Ben

[robnyc]

Regarding bias pots, I often recommend either of two safety measures here.

The simplest is to set the desired idle bias, I use 1.2 to 2mv on typical 2N3055 amps. Disconnect one lead to the pot and measure it. Replace it with a fixed resistor as close to that value as practical.

The other approach allows for future adjustment in the case of transistor replacement. Adjust bias for a level that it multiple of the desired value (transistor idling a bit too warm) and connect a fixed resistor across the variable with a value that is equivalent to that too-high value should the pot go open. This will allow a range of adjustment and a degree of safety if the pot fails -especially during adjustment.

These were both common addendum's to service info during the late sixties. Now the questions is; why didn't the manufacturers do this to begin with? A lot of the ugly flameouts of both amp and speakers that scared the hell out of people could have been prevented with one simple resistor per channel.

RobNYC