Come with me as I repair a Sansui AU-317 integrated amplifier and discuss problems associated with the wrong approach to caring for beauties like this.

This isn’t really a rant, we don’t rant here at Liquid Audio, but I do occasionally get frustrated, and it’s a chance for me to express my thoughts, to hopefully educate, inform and help readers make better decisions. Welcome back!

I’ve just repaired this lovely Sansui AU-317, a classic integrated amplifier from the late ’70s that we will chat more about as we go. This case presents a useful learning opportunity and provides pause for owners of such equipment to carefully evaluate who they entrust it with.

By the way, I’ve been busy updating the FAQs and the Hall of Shame, in particular case #16, featuring one of the worst-made hi-fi products I’ve seen in a while and a business owner who threatened me with legal action for making people aware of it. But I digress… Time to geek out a little bit at one of the best budget amplifiers ever made. Let’s go. 😉

For more about this story and a nice little video summary that has gone viral, by my YouTube channel standards at least, check this out:

The Sansui AU-317

Many of you will have seen my own Sansui AU-317 in the workshop over the years. I love these amplifiers, their baby brothers the AU-117 and AU-217 and the bigger brother AU-517 and AU-717, AU-D707/AU-819 and AU-919. I repaired an AU-217 only last week, and there is a lot to love about the slightly more capable AU-317. The rest of the range is, of course, legendary, and I love these stylish, serviceable and great-sounding amplifiers.

The Sansui AU-317 is a discrete component, full-featured design that can form the heart of a nice little system. With 50 Watts per channel and a solid 10kg build, these babies from 1977 – 1980 make modern $1000 amplifiers look like the toys they are. The AU-317 has a really good discrete phono preamplifier containing some very nice parts built in, a headphone ‘amplifier’ (really just a socket, but you know what I mean), and bypassable tone controls.

These amplifiers are completely repairable even now, more than 40 years on. Virtually nothing would render an AU-317 unrepairable unless it caught fire maybe, so there’s a lot to love here. Try repairing a Cambridge Audio CXA61 in ten years and you’ll appreciate my enthusiasm for equipment like the AU-317.

Sansui AU-317 Specifications

Courtesy of HiFi Engine (absolute legends, BTW!)

• Power output: 50 watts per channel into 8Ω (stereo)
• Frequency response: 5Hz to 70kHz
• Total harmonic distortion: 0.03%
• Damping factor: 70
• Input sensitivity: 20mV (mic), 2.5mV (MM), 150mV (line)
• Signal-to-noise ratio: 77dB (MM), 100dB (line)
• Channel separation: 65dB (MM), 73dB (line)
• Output: 150mV (line), 1V (Pre out)
• Speaker load impedance: 8Ω (minimum)
• Semiconductors: 51 x transistors, 25 x diodes, 2 x FET
• Dimensions: 430 x 110 x 340mm
• Weight: 9.5kg
• Year: 1978

The Right Approach

Liquid Audio focuses on preserving and rejuvenating classic hi-fi equipment, so I don’t want to waste too much time discussing what not to do, but there’s no avoiding it in this case. Regulars will know I have a particular dislike of poor workmanship, unnecessary work, guru-BS and recappers, so it’s probably no surprise that I’m going to dig into this a little here.

There’s a ton of Sansui AU-317s around, and they fail occasionally, in different ways, but repairing an AU-217, AU-317, AU-517, AU-717, etc., should be no big deal. Successfully repairing equipment like this is heavily dependent on a technician’s approach. Are ‘right’ and ‘wrong’ approaches? Absolutely, and it’s worth breaking this down so people understand it.

The wrong approach, so often applied, typically shares most of the following:

• Poorly skilled individuals unable to assess/investigate equipment faults/failures
• The need to seek advice in forums and buy parts on eBay
• Dodgy diagnoses, as we saw with this AU-317
• Shotgun parts replacement and ‘improvements’ suggested by armchair experts
• The absence of a skilled technical assessment to determine what is actually needed
• Removal of good original parts and replacement with cheap/eBay-sourced components…
• Such as parts ‘kits’, with no traceable provenance, aimed at shotgun repairers
• A lack of investigation, testing and test equipment and much guesswork
• Cheap tools, poor rework, lack of investment in expensive but necessary tools and equipment.

This approach is wasteful, largely ineffectual, prone to introducing new faults as the ‘Sansui specialist’ mentioned above did, and devalues otherwise perfectly equipment. This is also the most common approach because it requires the least knowledge, skill and technical proficiency.

There are some classics, even from here in WA, like a so-called ‘Sansui specialist’ who’s stolen from my website, asking in a forum about how to fix an AU-101 he’d broken. An AU-101 is one of the simplest amplifier designs, and yet these folks often cannot fix them without help.

The right approach is rare and very different, and:

• Is a conservative, precise, technically-focused one
• Honours the original design engineers above all else
• Focuses on assessing and caring for the unique needs of a piece
• Avoids soliciting advice about what poorly informed individuals think is ‘needed’
• Replaces parts conservatively, where necessary
• Utilises better than original parts, never Suntan capacitors or parts that don’t fit
• Installs parts precisely, more neatly than the factory original
• Focuses on repairing faults and restoring perfect operation
• It is informed by a technician’s extensive experience, knowledge and skill, and the condition and faults of equipment

This approach is rare because it requires an unusual level of experience, skill, technical knowledge, and the ability to find and solve sometimes tricky problems on one’s own. The upside is that caring for equipment in this way adds value, improves performance and increases enjoyment of it. It also builds invaluable experience and skill.

“She Needs a New Flux Capacitor, Sir…”

So, this Sansui AU-317 went to a local repairer via her previous owner. Like some others, this business has emulated aspects of my approach, content, and style, but unlike Liquid Audio, this business has changed its name at least four times over the last few years to dodge some appalling reviews and a significant case they lost in Joondalup Magistrates Court in 2024. The details really matter.

Be wary of businesses that constantly change names; the solar industry is a classic example. Successful, trusted businesses don’t change names, for obvious reasons, and I won’t be changing the name of my business any time soon. It is almost always done to remove a tainted business profile, thereby scrubbing negative reviews, feedback and allowing a fresh start. It will always be dressed up as something else.

How much this matters depends on the individual, but it matters a great deal to my customers and me, and I’m certain most folks would want these details before deciding where to take their cherished hi-fi equipment.

The amp went in for presumably the same issues she came to me for: unlistenable levels of distortion and DC offset that rendered the amplifier completely unusable. She was sent back with those same problems but minus all of her original high-quality capacitors and with a new set of crappy ‘Suntan’ brand caps, driver devices and trimmer potentiometers installed, at the owner’s expense.

That owner finally gave up on her after reading the diagnosis I’m about to show you and listed her for sale. The current owner, my customer, purchased her for next to nothing. The seller kindly suggested that the amplifier be brought to me, as he thought I would be able to fix it, which I did. Thank you to the previous owner.

What caught my attention, though, was the diagnosis the previous repairer provided, and I hope he won’t mind me quoting him:

“The transformer related to the power section is on its way out” and “the JFET ICs need to be replaced.”

Previous repairer

Riiiiiiighttt. Now, this is nonsense, and while the customer might not understand what it means, any competent technician reading this will think: “Hang on a minute, what..?!” The diagnosis is neither accurate nor possible in this amplifier. The dude might just as well have said, “She needs a new flux capacitor!”

Nonsense Breakdown

I think we need to dig into this nonsense because it’s unlikely anybody else will, and it’s potentially informative for readers.

The Transformer

What does “The transformer related to the power section is on its way out” mean, exactly? Is there a transformer party somewhere? Transformers – more than meets the eye! For the record, there is only one transformer in this amplifier, not a separate one for “the power section”.

Transformers either work or they don’t, so a transformer cannot be “On the way out” either. The only positive here is that The Power Section is a cool band name, though similar to another great band name… RIP, Robert Palmer.

The JFET ICs

Regarding the “JFET ICs”, there are no integrated circuits or ICs in an AU-317; it’s one of the lovely things about the design. The service manual refers to the input differential JFET pairs as ICs, but that’s a translation glitch; these are dual JFET packages, not integrated circuits.

The JEFT packages work perfectly, a fact confirmed by a quick functional test of the amplifier. Even if these JFET packages did “need to be replaced”, replacement is straightforward, so why not just do it then? Better still, why not test them and definitively determine whether they need to be replaced?

The Capacitors, Drivers, Trimmers

Lastly, why replace all the trimmer potentiometers and driver devices when these would have been perfectly good, based on working on dozens of AU-317s over the years? They were the original high-quality parts that did not need to be replaced. Likewise, installing Suntan capacitors is like removing partially worn Michelin tyres and installing a new set of cheap Chinese ones. It is an ill-informed, value-reducing, shotgun approach.

The reason for all this is a lack of experience and knowledge. The latest claims of decades of combined experience under yet another new business name are meaningless if the experience is spread across multiple individuals doing work like this. It could be a million years of combined experience; it makes no difference whatsoever.

Ethics & Integrity

OK, so the previous tech trying to fix the amp couldn’t do it. They likely felt they needed to tell the owner something useful, but couldn’t do that either, so they made stuff up. I understand, but that’s not good enough. Time and money-wasting practices like this have no place in this industry, and this is something I’m actively seeking to improve.

We have all faced repairs we couldn’t complete because time or a customer’s budget ran out, a fault was elusive, or because a mission-critical part couldn’t be sourced. This work can be difficult, but sending equipment back with misleading or made-up diagnoses and a bunch of junk parts is unethical and unacceptable.

Repairs may not be possible within the time or budget allocated, and that’s OK, as long as this is carefully explained and options are provided. It happened to me with a Marantz 6300 turntable: I was unable to repair it within the initial hour I billed for, and explained this. I noted that, given more time, I could repair it, but I wasn’t sure how long I might need. We’re all human, and this happens in all fields of technical work.

For anyone still wondering why any of this matters, I see two reasons:

• This amplifier was someone’s pride and joy, and they no longer own it because of what happened
• Unethical/unnecessary work harms equipment and the industry itself. I believe people ought to know about it.

I’m a stickler for doing things properly, and I’m not for a moment suggesting that I’m the only person doing sound technical work. Those who do are certainly in the minority, though, and so it’s immensely helpful to know who those individuals are.

The business responsible has since changed its name again and is still doing work like this, so here’s my challenge:

Rather than copying my ideas and content, which is easy, how about tackling the important stuff, like education, integrity, skill and professionalism?

These are harder and require more work, but generate goodwill and bring customers back. You’ll feel better about not constantly generating BS and disgruntled customers, and having to come up with yet another business name.

Service & Repair

OK, with that out of the way, let’s continue. The path to repairing this lovely classic AU-317 was to isolate, diagnose and repair the faults, of which there were two. That took six electronic components and some time.

Fault 1 – Distortion. The distortion was caused by blown current limit resistors in the bias/drive circuit that turned this amplifier into a class-B design with no driver stage. That means that huge signals would be needed to drive the output stage at all, and what you get would be badly distorted. I replaced the current limit resistors, set the bias current and bingo, the distortion was gone.

Fault 2 – DC offset in the right channel of the phono preamp. DC offsets are typically caused by leaky or drifty small signal semiconductors and the obvious pair to check in the right channel of the phono preamp is the right channel input differential pair. Sure enough, one device was bad, so I replaced both with a new hand-matched pair with improved specs.

Some critical cleaning, service and adjustment completed the work here and is what had this AU-317 running perfectly again.

Results

This Sansui AU-317 repair was particularly rewarding because her owner is such a nice guy and he appreciates the approach I’ve described in this article. He owns many Sansui amplifiers including a Sansui AU-101, AU-217, AU-555A and AU-4900, all of which I’ve serviced and repaired for him but his favourites are the AU-217 and AU-317 I recently worked on. Here’s what he told me about the AU-317:

[The] 317 sounds spot on, you are the man, thanks again Mike. She sounds bloody awesome, I hate to admit its in a tight second spot for my favourite.

Michael B

In terms of overhaul and restoration, less is more, especially with pieces in generally good condition. Overhauls by the right people are worth having done, but as always, the devil is in the details. The right technician won’t be getting their information from forums, installing Suntan capacitors or charging $1500 AUD to overhaul a basic amplifier. They will already know what to do, based on the condition of the equipment in question. Keep this in mind.

If you want to spend that much on a vintage amplifier, I suggest you take the $1500, sell the AU-317 and put all of that money into a nice Sansui AU-717, AU-819 or AU-919. That’s going to give you a far better piece of gear and bang for your buck than going crazy with an AU-317 or AU-4900.

The Bottom Line

Want a beautiful-looking, great-sounding, affordable, reliable, repairable moderate-power integrated amplifier? I highly recommend you consider the Sansui AU-317 if so. These are solid, well-designed, well-made pieces of Japanese hi-fi history.

Nothing like this will be made again and at their current prices, you can’t go wrong. I’d take one of these over new plastic rubbish with Bluetooth integration every day of the week. Just take it to someone with the right approach when it needs service!

As always, thanks for reading and I hope you found this article informative. If you’d like me to look at your Sansui AU-317 or any other lovely Sansui amplifier, get in touch.

Sansui AU-317 Integrated Amplifier

$500 - $1000 AUD

8.4

Build Quality

7.5/10

Features

8.5/10

Sound Quality

7.0/10

Servicability

9.5/10

Bang-Per-Buck

9.5/10

Pros

• Full-featured integrated amplifer
• Excellent phono preamplifier
• Tape and headphone functions
• Superbly servicable
• A great performer in its niche

Cons

• Lacks slam, air, refinement etc. of course, given what it is.