Japanese SE stereo amp

[retailer]

I Decided to make a separate post about this SE EL84 amp I acquired a few days ago, a length of masking tape stuck to the side of the cabinet read 'both channels distorted Sept. '16 - it is a rebadged Japanese made mid 1960's integrated amp. A long search of the internet turned up a number of amps that look identical but with differing names - Planet - Linmark and AMD, but sadly no circuit. Removing the bottom panel reveals that it has been worked on - this can be good if the repairer knows what they are doing and bad if they just think they know what they are doing, without a circuit there is really no way of knowing if correct components have been used - if the message on the masking tape is correct - both channels distorted then the most likely cause is power supply issues as this is common to both channels.

[teetoon]

Hi. This is probably better known here as an Eagle SA 200.
Lots of info about those can be found with a web search and the first one I looked at mentioned all the coupling caps were leaky. Not surprising for an amp of this age.

Hope you can find a circuit. They seem to have been well regarded.

David.

[Craig Sawyers]

From the images it looks like all the electolytics have already been replaced, including the reservoir capacitor in the power supply.

Craig

[trobbins]

Further to the Oz distributed brand names - I've seen Rapar, Star, Muse and Encel (which I have a X1212 model which was introduced circa 1964). Often they had a similar SA-xxx model number.

The STAR SA-30 manual is on-line (Hifi Engine) with schematic, although it has a 6GW8 PP output stages.

[retailer]

Had some time to spend on this amp today and using an old mobile phone as a music source, I found it is indeed distorted on both channels - the distortion getting progressively worse as the amp warms up - I decided to pull all of the preamp valves, feed a sinewave straight into one of the EL84 grids and scope the output while on a dummy load - my dummy load is a cast Al box with switched 4,8 and 16 ohm resistors - with the dummy load connected to the 8ohm tap the output was very low, less than 1watt at onset of clipping - switching the dummy load to 16 ohms while still on the amp's 8ohm tap bought the output up to about what it should be approx 3.5watts. I don't have a circuit, but it looks as if the wiring from output transformer secondary is all original and hasn't been messed with - I checked the output transformers by running 20V ac through the primary and measured the secondary voltages - it all checked out.

A pair of large 200 ohm resistors are in the output transformer secondary circuit so I traced it out and found a strange (to me) speaker/headphone switching setup, the 8ohm tap is grounded rather than the common connection - a large 200ohm resistor is permanently wired between the 16ohm tap and common - a DPDT switch selects headphone or speaker to the common, the 200ohm resistor is probably there as a safe guard 'load' just in case headphones are selected but not plugged in.
I can't see that the 200ohm resistor would cause any issues but when I get some more time I'll rewire it in a more conventional fashion with the common as gound and disconnect the 200ohm resistor to see if makes any difference, I may have to play around with the value of the NFB resistor, it's currently 5k.

[roffe]

A circuit diagram for Linmark:
https://antiqueradios.com/forums/vie...?f=13&t=198932

[retailer]

Originally Posted by roffe

A circuit diagram for Linmark:
https://antiqueradios.com/forums/vie...?f=13&t=198932

Many thanks for that - I spent a good number of hours on internet searches and came up with nothing.

[Valvepower]

Hello,

Interesting circuit. The output stage and tone controls look quite orthodox although it’s interesting the use of an ECC85 in the driver stage (V3) for the EL84.

Maybe I'm losing the plot here, and EQ (50K and 0.001uF in V1+2b grid circuit) is applied to the Aux, Mic, Tape, Tuner inputs as well as the phono inputs. This will give a little LF lift, although I can’t see any HF EQ – maybe one just turned the treble down for MM HF EQ on phono input and the Bass down on the line inputs to negate the bass lift!

I figure there is quite a bit of gain in the preamplifier V1+2 and personally, I would be inclined to remove the EQ network and loose some gain in V1+2 for modern line levels. I’d also be inclined to use an external phono amplifier if one is using it with a MM cartridge, again removing the EQ network and loose some gain in V1+2. It may only need one of the two stages.

Or equally leave as-is and live with the high input sensitivity and light bass lift.

Edit… I have seen some budget amplifiers (Stereosound ECL82 amplifier) from the 1960’a with bass lift on all inputs.

Anyway, that's just my two penn'orth.

Terry

[PJL]

Is the distortion the same on both channels? Looks like most of the caps have been replaced and it all looks very tidy too.

I would do a few voltage checks on the valves before messing with the feedback.

[Valvepower]

Sorry, I got so tied up with the circuit...

I agree with PJL and do some voltage and similar checks before messing with the feedback.

I must admit to not taking replaced parts as being OK, as I’ve had instances where newly replaced parts being faulty.

I think the feedback comes from what could be 32Ω tap as I can just make out what I think is 32Ω on the A-Channel. The feedback comes from the 8 and 32Ω winding where the 8Ω tap is earthed.

The 200Ω gives a load if the headphone switch is switched to headphone with no headphones plugged in. The 8Ω tap is connected to ground so be careful connecting to the ‘0’ socket on the speaker connector with a piece of kit that is also earthed to the chassis as you could potentially short the ‘0’ to the 8Ω tap.

Terry

[retailer]

Distortion is the same in both channels, however I only tried the aux input which I expected at the least to be similar to the tuner input, this is where I made a mistake by not tying other inputs - with my music source connected to tuner input, it sounded much better and more like I had expected, looks like the first preamp stage is easily overloaded. I took a few voltage measurements and the anode of the first 12AX7 triode is at 80 volts, this is lower than I would have expected - the second triode anode voltage is 175 volts.
The mains lead needs to be properly secured - the exposed terminals on the fuse holder need to be insulated and the remaining original capacitors need to be replaced - I checked a few and they all show a bit less than 1M resistance at 250volts on my insulation tester - I'll also add some sort of attenuation to the aux input to bring it somewhere close to a line level input, it's tempting to go all out on modifications, but it is a budget no name amp - if it works well enough for use as a workshop amp for now I'll be happy.

Yes - a word about that strange (to me) setup where the 8ohm tap is earthed, this is a bit of a trap I feel when it comes to connecting a scope to the output - scope probe ground clip on common and scope probe tip on the 8 ohm tap effectively shorts the output unless the amp or scope is fully isolated from the mains ground, zero signal would show on the scope - one could easily be fooled into thinking the amp had a major fault.

[retailer]

Just had a thought - might experiment and connect the tape out sockets directly to the second stage preamp - tape out is seldom if ever used these days.

[retailer]

I moved the coax coming from the tape out socket to the point marked with a red X and gave the amp a short test with the same music source, an old Samsung Galaxy 2 - volume was good, but I thought it sounded slightly different, so I tried a bit of an A/B test moving between the Tape Out and Tuner and sure enough the Tuner input sounded slightly muddy - curious about this I decided to do a very simple sine wave frequency response check into a dummy load, and scope the output, I wasn't expecting anything great so I chose the -3dB point as the minimum. With the Tuner as the input the -3dB point was at approx 10Khz, not a great result and worse than I thought it would be, with the Tape Out(which is still connected to the point marked with a red X) it was a bit better at around 12.5KHz, at the bottom end I was surprised to see with both inputs the -3dB point was at approx 25Hz which is half decent given it was budget amp.

In checking the response with the sinewave input at the EL84 grid and also at the volume control pot, the -3dB point was somewhere in excess of 25KHz, so something wasn't right in the preamp section. There are still original paper capacitors in the pre-amp and tone control which I know are leaky - I've ordered some replacements and if this doesn't improve the performance of the amp then I'll look at rewiring the preamp - rather than try and design something from the ground up I'll just copy a circuit from a better quality Japanese amp, I was looking at a Pioneer SX34 a few days ago and this looks like a good starting point, it should allow the use of the original function switch.

[trobbins]

The 'strange' speaker output connections is likely related to avoiding resonance peaks from the output transformer. The choice of output transformer tap for grounding can have a significant influence. The 200 ohm loading may also be mainly about alleviating resonance peaks, as the feedback loading itself would probably avoid over-voltage issues from no loading.

One advantage of setting up a soundcard with spectrum analyser software as a bench tool is that it makes fault-finding and assessing performance like frequency response and resonances a breeze for amps like this.

[retailer]

A update on the little Japanese EL84 SE amp - the capacitors arrived and after replacing the rest of the original paper capacitors, I ran a few tests and noted that the frequency response hadn't substantially changed, probing with a scope and signal gen I narrowed the offending section down to the tone controls, and found that one gang of the treble control had a linear rather than a log taper and it was this channel I was testing - switching the testing to the other channel I found the frequency response was quite acceptable, I'll probably leave as is, most people tweak the tone controls to make up for speakers/rooms etc anyway.
More important than the tone control is the very high input sensitivity, causing the first preamp stage to overload, I tried a cd player into all of the inputs in turn and they all produce bad distortion. Again with a scope and signal gen I found the maximum input voltage to be less than 10mV before the ouput of the first triode in the preamp starts to clip.
I've decided to rewire the preamp, I can't do anything too fancy as I'm stuck with a simplified input function switching setup - I'm going to more or less copy a Pioneer SX34 pre-amp circuit, the mag phono input will be direct into the first triode and then routed to the second triode via the function switch - the other inputs will bypass the first triode and be routed via the function switch to the second triode Once it's wired up I'll run a signal gen into the inputs and get an idea of the signal level needed for the amp to have full 3.5W output with the volume control at the 3o'clock position.

[retailer]

Started work to rewire the front end of the amp and I came across these two tucked under some shielded cables, looks like someone was asleep on the job and missed making the solder joint, the capacitors are part of the loudness switch, the circuit I have must be for an earlier version as it doesn't show a loudness switch - something to attend to after the rewire.
I completed the rewire and it looks to be a success, lugged a turntable out to the workshop and tried the mag-phono input, cartridge is a Sure M95ED, don't know what this cartridge output is but low-level unobtrusive volume through a pair of Realistic Minimus 7's was at around 11 o'clock on the volume control and at 3 o'clock it was far too loud for the workshop. A simple resistive voltage divider was wired across the tape input to reduce the input voltage by 45% and with my CD player, volume levels were on par with the phono input - volume at 3 o'clock was far too loud for the workshop, sound quality was as good as I expected, I'm quite pleased.

[Analogue man]

Now to dig my Eagle out and go through that to get it working.

[GroovyG]

Originally Posted by retailer

Started work to rewire the front end of the amp and I came across these two tucked under some shielded cables, looks like someone was asleep on the job and missed making the solder joint, the capacitors are part of the loudness switch, the circuit I have must be for an earlier version as it doesn't show a loudness switch - something to attend to after the rewire.
...

Others here will know more and know better, but I recall another amp under discussion here that had similarly half-attached components near the inputs, provided for "trim to suit in situ" purposes. Perhaps these caps are similar.