Altec Lansing 1594A - 1967 [100W Solid State Mono Power Amplifier] BOM + Manual

[TelRay]

thx Leo!

yes, I am lucky to have a VARIABLE ISOLATED TRANSFORMER and when installing new MAIN / BIG CAPS I always start with no load, volume to zero and set the transformer to 40 VAC. Leave it there for about 30 sec and then slowly increase to 80 VAC for another 30 sec and finally to 120 VAC.

Exactly that, I have a couple of 8 Ohm loads

Your step by step description is reassuring and I appreciate it!

I have looked "crossover distortion" up to make sure I know what I am looking for in the scope (attaching for future reference as many inexperienced people could find it useful)

[TelRay]

below my findings,

Got the units warming up for 15 minutes and the transistors were warm but not hot (can touch them for more than 5 seconds). Q11 and Q12 being the hottest ones in both units.
set up the PC to generate a 1,000 Hz sine wave and got the output adjusted to 0.8 V (peak to peak, measured with the scope)

On UNIT#1 (the "old one" still at Rev 4) I was able to go to full VOLUME without seeing any distortion (either peak or crossover).
I got the unit delivering 18.3 V and drawing 1.1 A from the mains
Disconnected loads, input and brought the volume to zero and hooked the MMT across the speakers output Measured 0 mVDC
On idle mode the unit was pulling 0.6 A

On UNIT#2 (the "new one" at Rev 54) I was also able to go to full VOLUME without seeing any distortion (either peak or crossover).
I got the unit delivering 20.0 V and drawing 1.2A from the mains
Disconnected loads, input and brought the volume to zero and hooked the MMT across the speakers output Measured 0 mVDC
On idle mode the unit was pulling 0.6 A

I played with the bias adjustment on both units and saw the waveform growing and the current draw going higher but not seeing distortion on the scope.

I think I might be missing something because I should be able to get these things to show some distortion at some point, right?

[Helmholtz]

Input sensitivity is specified as 0.8Vrms. This corresponds to 2.26Vpp.

[g1]

I'm not sure which service manual you referred to about the bias adjustment on scope.
The 1594A schematic you posted in the first post seems to show only a 'balance' control that I can see. From that I gather you may not be able to introduce crossover distortion.

The 1594C manual from post #3 shows 2 separate bias adjustments so you can set both bias and balance. That manual refers to idle current specifications for the 2 halves (push & pull). That version I think you would be able to bias cold enough to see x-over on scope.

[TelRay]

Input sensitivity is specified as 0.8Vrms. This corresponds to 2.26Vpp.

right you are: RMS!
so the signal input is too low to trigger any distortion thya would allow visually adjusting the bias
vielen dank!

[TelRay]

I'm not sure which service manual you referred to about the bias adjustment on scope.
The 1594A schematic you posted in the first post seems to show only a 'balance' control that I can see. From that I gather you may not be able to introduce crossover distortion.

The 1594C manual from post #3 shows 2 separate bias adjustments so you can set both bias and balance. That manual refers to idle current specifications for the 2 halves (push & pull). That version I think you would be able to bias cold enough to see x-over on scope.

thx sir,
i always refer to the manual i posted for the model i have: 1594A
on page 3 they describe the adjustment of the bias voltage using potentiometer P2 (this is how they describe it under the chapter CONTROLS)
though they might be wrong

[Enzo]

It would be colder bias settings that increase crossover distortion, not hotter. Once the amp is biased hot enough to eliminate the Xover notch, hotter settings won't bring it back, they just make the mains draw run higher.

[Helmholtz]

right you are: RMS!
so the signal input is too low to trigger any distortion thya would allow visually adjusting the bias
vielen dank!

No need to increase bias (and adjust balance which would show as DCV at the output) as you don't see crossover distortion, which would show also at lower output level. But if you want to see clipping you would need to increase the input signal by around a factor 3.

[Leo_Gnardo]

It would be colder bias settings that increase crossover distortion, not hotter. Once the amp is biased hot enough to eliminate the Xover notch, hotter settings won't bring it back, they just make the mains draw run higher.

Plus, overbiased output transistors heat up more. Which can bring them to the brink of failure if they heat up too much. Beyond that, they fail, as regular junction transistors* do, by thermal runaway. In this instance more is not better. "Just enough" will do. Park 'em in the Goldilocks zone & you're golden.

* your typical NPN & PNP devices.

so the signal input is too low to trigger any distortion that would allow visually adjusting the bias

Not so. Even a small signal would show crossover distortion if the bias is insufficient. That kind of distortion makes signals sound "fuzzy" at any level. And most annoyingly, even with the volume fairly low. "Geeze it sounds like my speaker's blown or something." or "The music sounds like it's being filtered through gravel, yecch!" Like that.

Given that you've come this far successfully, how much does the heat sink warm up over the course of 15 minutes or so, with no signal going through the amp? If it's just a little warm, then later you scope a sine wave and it looks good right up to clipping, you're in the zone so leave it alone.

[g1]

1594A
on page 3 they describe the adjustment of the bias voltage using potentiometer P2

Ok, For the 1594A I see they are talking about clipping and distortion. I believe they mean if at full power, the top clips earlier than the bottom, or vice-versa. So you would adjust the 'balance' trimmer so they both clip at the same time, resulting in max. undistorted output.
If you run it up to just clipping on the scope, you will probably see that is the effect the trimmer has.

[TelRay]

thx guys! will re-read everything to the last detail to absorb as much knowledge as possible

i want to deviate the conversation a bit from only the crossover distortion (which is something i just learned about thanks to your contributions to this post). as it was meantioned by Leo before I should also look for clipping on the peaks, what I was intending to say (“my ingles no bueno!”) is that as Hr Helmholtz pointed out, the input signal might be too low to produce this kind of distortion.
So, even though I see no crossover or peak distortion, that doesn’t mean the amps biased (or balanced) correctly as long as i do not increase the input signal so it would actually triggers something. please let me know if my logic is OK and if I should try again with a hotter signal that will be 0.8 V RMS

TEMPERATURE:
when playing music Q11 and Q12 are the hottest transistors (both amps)
i will try to get a thermometer that can read that temperature

AMP 1: produces a low hum (audible only with the ear aginst the speaker, no change with volume). i moved P2 “by ear” until the hum was minimal

AMP 2: Q11 and Q12 run hotter than on AMP 1

[g1]

Q11 and Q12 are the output devices. When putting out signal into load it makes sense they would be the hottest.
You can measure the idle currents indirectly by measuring DC idle voltage across R40 and R50, then calculating with resistance value to solve for current. Then you can see how much hotter the one amp is idling compared to the other.

AMP 1: produces a low hum (audible only with the ear aginst the speaker, no change with volume). i moved P2 “by ear” until the hum was minimal

Just like the 'output tube matching' control on SF Twin.

[Leo_Gnardo]

Q11 and Q12 are the output devices. When putting out signal into load it makes sense they would be the hottest.
You can measure the idle currents indirectly by measuring DC idle voltage across R40 and R50, then calculating with resistance value to solve for current. Then you can see how much hotter the one amp is idling compared to the other.


Just like the 'output tube matching' control on SF Twin.

What G1 said ^^^. If the output transistors get hot without any signal, then it's a good choice to reduce bias current. You can input a sine wave, watch the 'scope, see when that crossover notch appears, then back off a scoche to where it disappears & park it there. Heck in a pinch I've done it by ear, no scope, that can work too. When the annoying overtone stops, dial in a wee pinch more and Bob's yer uncle. A dot of fingernail polish on the trimpot to hold it in place is a nice final touch. Especially if it's some unusual color - I got some flourescent orange - you can't miss seeing it.

Once you have 'em dialed in, play some music thru your speakers at a level you'd ordinarily listen. Or even louder if nobody's around to be bothered by it. If the output transistors get steaming hot after say 15 minutes of brutal thrash rock or whatever, consider getting a fan behind each of those amps to keep 'em out of the overheat zone. BJT's - ordinary transistors - fail when overheated. It gets into a runaway situation and all too often super hot transistors short without any prior warning, so keep 'em kool especially under prolonged high power use.

[Enzo]

I found a quick trick was to monitor mains current. Start teh bias at coldest and slowly turn it hotter. At the very point the minimal mains current starts to rise. Stop, and I do back it off a tiny bot. The reason teh mains current starts top ramp up is that you have reached the point both sides of the push/pull have started to conduct - passing current from rail to rail.

[TelRay]

I found a quick trick was to monitor mains current. Start teh bias at coldest and slowly turn it hotter. At the very point the minimal mains current starts to rise. Stop, and I do back it off a tiny bot. The reason teh mains current starts top ramp up is that you have reached the point both sides of the push/pull have started to conduct - passing current from rail to rail.

yes! I've noticed that when keeping an eye on the Current indicator needle on the variable transformer moving the potentiometer from one extreme to the other will make the current draw to go down and then go up again. I thought it would be logic to leave the bias balance at the lowest point too