Testing an amp at full voltage, but limited current...

[df_genius]

I have a problem with my Marantz PM17 KI, I think I am very close to fixing at it works great when connected to a 60W bulb, and the bulb is nice and dim. Music plays and it sounds great. I have had a real problem getting it back on the road after being over-zealous with the bias adjust a couple of months back.

However, when I connected it straight to the mains last night the outputs blew straight away. My theory is I would like some way of limiting the current to prevent the outputs from blowing but run it at full voltage to be able to pick the problem up.

Any ideas?

[Zedhead70]

I am not familiar with the PM 17 but I've repaired my share of PM94's, I generally use a 100-150w light globe and having had your problem occur on more than one occasion I check as much information as I can before going " live". Are the supply rails going full voltage on the 60w globe? Is the problem with both channels? If not, use the good channel to compare with the bad channel.Check the bias even on the test globe as this can point to problems. it won't go to full bias but if there is an imbalance between good and bad channels it may show something. Is there any DC on outputs above 30mV? Look at the voltage on the bias transistors they can sometimes give clues.Good luck! Regards Craig.

[df_genius]

I have tried both 60W and 100W bulbs.

It is just the left channel, the right is OK. All checks I have made now measure both left and right channels the same. All I found last night was a 100Ohm resistor (near bias pot but not close in the circuit) that measured correctly but was marked, I am hoping that is the culprit!

Bias from cold and with the 60W bulb was 35mV and within 1mV of each other, DC offset I had adjusted to less than 1mV on both channels.

I think tonight I will leave it on for a couple of hours on a 100W bulb.....

[EchoWars]

Do not set the bias with the lamp in place...except to set it to some very small figure (like 1mV or something). Set bias ONLY with full voltage supplying the unit.

If that still results in blowing fuses and output devices, you better look at buying your transistors somewhere else. We have big problems with counterfeit transistors in the US, but from what I hear, nothing like what is happening in the UK.

[df_genius]

The only thing I have done to adjust the bias was to turn the pots right down.

This evening has been fun (!). But I think I have made progress. I found an error in the manual and found mysel replacing two 680 Ohm resistors for two 330 Ohm devices

EW - I have sourced the transistors you reccomended and they work fine, having installed them in the good (right) channel and everything is OK.

Just with the right-channel board in, with no volume etc. my 100W bulb is dim to the point of being off, and the amp gets 232V at its input terminals. It works fine.

With the left channel board in (on its own) the bulb is reasonably dim, i'd say at about 10% brightness. BUT still much brighter than just having the right-channel board in.

Running accross the centre of the board, there are three 330Ohm ESD/ESP resistors in series (can't rememeber excatly what the manual said and it's in the garage), one end of which is connected to the centre pin on the emmittor resistor (aka 0V?). On the right channel these are at 0V DC wrt Ground. When I touch the contacts with my meter probe it causes the relay to kick out and back in. Obviously a sensitive circuit.

With the left channel, the resistors seem to be between 0.6 and 0.8V (the 2nd and 3rd away from the emittor resistor) above ground. Touching them with my meter probe causes the bulb to get bright quickly then die down.

My head tells me that the outputs are working 0.6-0.8V above zero for the circuit and that is causing the problems, but I don't know where to look next.

I have just been over all resistors, diodes and transistors and done basic testing (no power though). Testing this powered up is not easy as the boards are on their side, facing the transformer/power caps about 2 inches away

[d3imlay]

Look for an intermittent bias pot. If the bias pot opens, does the bias go full on or full off? Also look for cracked solder joints. There should be a diode or transistor mounted on the heatsink for thermal stability. Look for broken leads. Once you think you have the problem fixed, blow a hair dryer on the thermal tracking device from about 12 inches and watch the bias. It should begin to drop as the heatsinsk & xistors warm up. This should take less than 1 minute.
Are you using exact replacement outputs or subs? If subs, be sure they meet or exceed the safe operating area of the original devices.

[df_genius]

If the bias pot opens, it goes full-off. I have tested the bias pot and can't find anything wrong, it is a 2K Ohm pot, I think the fault lies elsewhere, on one end of these 330 Ohm resistors. Any ideas what this circuit is for?

Is it worth taking the cover off the HDAM and testing the parts in there?

One thing I am a bit worried about is that the circuit board seems more delicate than I am used to working on older kit. I have ended up extending the legs of a couple of components to reach to the next connection. I am trying to be careful!

[jblmar]

Check the diodes that are connected to the fuse that blows. You may have a shorted or leaky diode.

Ron