Focusrite Saffire Pro 40 re-cap

[Khron]

Re: Focusrite Saffire Pro 40 re-cap

Makes a person wonder how the "stock vs. stock" comparison might've went...

[12Bass]

Re: Focusrite Saffire Pro 40 re-cap

Anyone care to provide a stock Impact Twin?

I'd guess that there would be a significant differences in sound between a stock Saffire Pro and Impact Twin due to basic design differences. I did make some subjective A/B comparisons as I modified the two preamp channels on the IT. For example, I compared one modified preamp channel with polypropylene bypass capacitors in parallel with the electrolytics in the signal path with another without added bypass caps. The channel with bypass caps was noticeably clearer, with more apparent subtle detail. I was somewhat surprised at the difference. Electrolytic capacitors have high ESR at high frequencies and much higher distortion levels than polypropylene. In my experience, bypassing electrolytics with film capacitors in the signal path makes a significant difference in overall transparency. Thus, I added polypropylene bypass caps to all of the signal path electrolytics in the Impact Twin.

On a side note, my Saffire Pro 40 shows quite a bit of "wavy" low level distortion when analog loopback testing with RMAA, which I think may be due to insufficient power supply filtering. I may try replacing the supply electrolytics and see if that reduces the distortion. Perhaps the stock capacitors are failing or poor quality? My Impact Twin also shows some spurious high frequency noise in RMAA, perhaps due to the internal switching supply?

[Khron]

Re: Focusrite Saffire Pro 40 re-cap

Perhaps you should start your own topic with the Impact Twin mods then, perhaps?

"High ESR at high frequencies" - measured that yourself, have you? And how can one quantify (if not measure) transparency?

The description of "wavy" low level distortion might be better described with a screenshot, maybe?

[12Bass]

Re: Focusrite Saffire Pro 40 re-cap

Perhaps you should start your own topic with the Impact Twin mods then, perhaps?

"High ESR at high frequencies" - measured that yourself, have you? And how can one quantify (if not measure) transparency?

The description of "wavy" low level distortion might be better described with a screenshot, maybe?

I've read about the characteristics of various types of capacitors and have seen test results. Electrolytics are among the worst for the signal path. Transparency could probably be quantified and measured somehow, however, what I meant was subjectively audible transparency. As mentioned previously, I did perform RMAA analog loopback tests on my Gina24 both before and after modifications and that testing showed a very obvious decrease in various types of distortion (notably THD and IMD).

Pro 40 RMAA screen capture:

[Khron]

Re: Focusrite Saffire Pro 40 re-cap

FYI, those peaks at 60 / 180 (3rd harmonic) / 300Hz (5th harmonic) are somehow mains-frequency related. Although, at -117dB and lower, the audibility of even those, is... debatable, to say the least

PS: http://sound.whsites.net/articles/capacitors.htm

<slightly mean jab / exaggeration> What next, we need certain "special" cables for the "best" sound?

[12Bass]

Re: Focusrite Saffire Pro 40 re-cap

Some capacitor info here: http://www.reliablecapacitors.com/pickcap.htm

I had a link showing oscilloscope traces for various dielectric types but cannot find it at the moment. Will post it later if I can find it.

As far as cables go, yes, they can make a significant difference in sound, particularly cable capacitance, when used with something like an electric guitar or bass where they interact with the pickups and alter the frequency response. These effects are easily measurable.

[12Bass]

Re: Focusrite Saffire Pro 40 re-cap

Here's the link showing the non-linearities of various dielectrics: https://web.archive.org/web/20120110...h102/caps.html

[Khron]

Re: Focusrite Saffire Pro 40 re-cap

Walt Jung is, of course, a well-respected engineer, but on the other hand, that article is from 1980 (nineteen-eighty), the best part of 40 (forty) years ago. I'd like to think capacitor technology hasn't entirely stagnated since then, y'know?

I'm not saying none of that is true, only that it's "a bit" out of date, i think.

And the second link is (only) about... dielectric absorption, is it? With no real / exact description of how that relates to the "sound", or what effects it translates into. Unless i totally missed something?

[Deep South]

Re: Focusrite Saffire Pro 40 re-cap

Khron,
I joined this forum some years ago thinking it might be very useful if I wanted sound advice on what I needed equipment wise in transferring my LPs to a digital format, life or rather illness got in the way,until now.

I saw this thread and found it interesting and not surprising that PSU cap changes make a difference.

Also noted your scepticism about audiophile/fool caps. there is a lot of b/s in the audio world and incredibly a complete disregard of sound science from way back.

Coupling caps have a bad name with many amp designers for a very good reason - instead of doing their job ie. to couple they interfere and 'colour' the signal, same goes for most resistors in the signal path.

Just after I was persuaded by a Dutch friend to shell out for some seriously expensive hand made silver foil coupling caps I discovered the Russian mil.spec. K73-16 caps.

There's a thread all about them on diyaudio, apart from a couple of androids who have never used them but 'know' they are no good, only one actual user did'nt like them.Jung slagged of the materials used to make the K73s They are 'transparent' which is also the best way to describe Z foil resistors. If there is crap upstream either, of these components will expose it. The only negative for the Focusrite might be that they are axial style caps.

I have two hybrid amps, one a h/amp and the other an integrated. On the integrated the first cap after the shunted pot is a 0.1uF/250V cap, changed out a Wima for a K73-16 @ 20mm x 8mm. The next cap is a 1uFx 250V used the K73-16, transformed the amp completely. These caps aren't big, so I assume that the 16V ones will be quite small. The negative for the Focusrite modding maybe one of headroom since they are axials and would have to be soldered on end.

Whether the kind of values you want are still available, I don't know, what I do know is that they were dirt cheap. Z foils are'nt ,especially the SMD type but what a difference they make as well. An Alps Blue standard pot is a piece of crap but shunt it with Z foils and it's transformed.

Analogue i/connects - before the end of the 19th century Oliver Heaviside had produced definitive work on electrical conduction that has never been gainsayed to this day but the audio world is in total denial of his findings. I won't add to this because so many think i/connects are i/connects and that's it, same goes for RCAs and XLRs, plugs are plugs though there are idiots who spend fortunes on super duper ones.Just take a look at accepted wisdom on analogue cables and then study Heaviside's work.

I have nothing to say on digital cables as I have never looked at their materials or construction - yet. The most important component in the whole chain must be the signal transfer surely?

[Khron]

Re: Focusrite Saffire Pro 40 re-cap

Well, for the record, i already sold my Saffire Pro 40 over two years ago "anyway", so...

I don't suppose you have any sort of before / after measurements of your mods, do you?

And any links to the exact parts of Heaviside's work that you're referring to?

[12Bass]

Re: Focusrite Saffire Pro 40 re-cap

Also noted your scepticism about audiophile/fool caps. there is a lot of b/s in the audio world and incredibly a complete disregard of sound science from way back.

Coupling caps have a bad name with many amp designers for a very good reason - instead of doing their job ie. to couple they interfere and 'colour' the signal, same goes for most resistors in the signal path.

FWIW, I've done quite a bit of capacitor comparison testing and can definitely say that there are significant differences in how they sound. One particularly interesting example was in a modified distortion pedal. I tried ceramic, polyester, and polypropylene capacitors of the same value and they all sounded different, depending upon the application. For example, I found that a polypropylene capacitor used in a simple low pass filter sounded "smoother" as in it sent more of the harsh upper harmonic distortion products to ground than a similar value polyester cap. I was somewhat surprised by this and swapped back and forth a few times before settling on polypropylene in that particular part of the circuit. In other places I've found ceramic or polystyrene to work better. Generally speaking, I avoid putting audio signals through electrolytics if at all possible.

[h2charon]

Re: Focusrite Saffire Pro 40 re-cap

Hi folks - i´ve got a blown C3 on my saffire pro 40 pcb which made quite a mess all over. i changed the cap - couldnt get the original capxon cap and replaced it with a KXJ cap. still not working. i had already the last years issues on startup like others up in this treat with delayed startup. i am not sure if the blown cap is the problems symptom or its origin. any suggestions ?
best herb

[bobbykonders]

Re: Focusrite Saffire Pro 40 re-cap

Well, for the record, i already sold my Saffire Pro 40 over two years ago "anyway", so...

Hello,
I just found this post about upgrading the op-amps on the Focusrite pro40.
First of all, thank you very much for sharing this knowledge.
I want to do some mods on my Saffire56, very similar design, and i have few question about it:

Can I replace the op-amps without change/modify the PSU? I mean, there is enough current to feed them all once replaced?

Is replacing caps needed after upgrading the op amps?

My main idea is to use 8 x OPA1612 to replace the 8x njm2122 on the preamp section. Then try to replace few njm4565 only for the output sections. (main out 1-2) and the 8 ones on the pre-amp section (if it makes any improvement).

What do you suggest about it?

thanks very much

[bobbykonders]

Re: Focusrite Saffire Pro 40 re-cap

well I forgot to mention this: obviously the njm4565 will be replaced with LM4562.

[SolderFumes]

Re: Focusrite Saffire Pro 40 re-cap

I hope I'm not committing mortal sin by reviving an old thread on my very first post. If so, my apologies. However, this is the best Saffire Pro 40 component-level discussion anywhere. Hopefully Khron is still keeping an eye on this thread even though he sold his Saffire.

I bought my Pro 40 new in 2015 but haven't used it regularly until recently. Now it has developed a problem that I would like to remedy. After it has been on for about an hour the inputs will get VERY distorted. Extreme distortion with strange harmonics and zero sustain. This occurs even when the unit is not connected to a computer. Since it consistently happens after about an hour, I assumed it was a heat issue. To test this, when the distortion began to appear, I removed the cover and placed a small fan over the power supply and the section of the main board next to the power supply that holds the voltage regulators. I did this for about a minute and the distortion disappeared completely. OK, it is heat related. Also, if I leave the cover off, the distortion does not comes back. However, this really isn't a viable long-term option. Since this is a new development it seems that some component or components have degraded and are now more susceptible to heat. In researching this problem, I came across this thread and it is a gold mine.

I have read the whole thread a few times, hopefully I haven't missed anything.

First off, I want to re-cap the power supply and get rid of the C(r)apXon caps. It is possible this may solve the problem, but I'm not sure. I would also like to replace the two TO-220 packages on the power supply. The switching FET on mine (CS8N60F) is a slightly different part than what Khron had(FQPF8N60C). Both of the replacements suggested are now considered obsolete. I would love to know what good current replacements would be.

Since I will have to order parts, I thought I should plan to order other components in case the power supply isn't the source of the problem. Here is what I am thinking:

1. Replace both LM317T's and the LM337, and/or put heat sinks on them.
2. Replace all the caps on the main board that are in the hot zone around the voltage regulators and the two 22ohm 1-3 watt resistors. These resistors put out a lot of heat. Khron had 5 SamXon 220uF 50V caps here, but mine are C(r)apXons!
3. Replace the large 22 ohm resistors with higher wattage resistors, say 10 watts. Would those run cooler?

Hopefully someone knowledgeable can chime in and give me some feedback on my ideas and let me know if I'm on the right track or if I'm way off. I just want to get this working as it should and learn as much as possible in the process.

Thanks for looking.

[Khron]

Re: Focusrite Saffire Pro 40 re-cap

Caps will be your first bet. If they go bad, they can easily throw the regulators "off" and into instability or whatever.

Personally i wouldn't replace any of the silicon unless it turns out to indeed be at fault, AFTER the re-cap.

Why would you want / need to replace the switching FET, if it isn't blown? Doesn't need to be anything special anyway, both those part numbers are just bog-standard 8A 600V MOSFETs.

How would replacing those resistors diminish the heat they put out, of the ohm value is the same? Ohm's law still applies

If anything, you might wanna re-melt the solder joints of the three regulators, "just in case". Although if they got hot enough to have been a problem, you'd likely see some PCB discoloration in the area.

[SolderFumes]

Re: Focusrite Saffire Pro 40 re-cap

Hi Khron,

Thanks for your reply!

Caps will be your first bet. If they go bad, they can easily throw the regulators "off" and into instability or whatever.

Personally i wouldn't replace any of the silicon unless it turns out to indeed be at fault, AFTER the re-cap.

Great. That is what I will do.

Why would you want / need to replace the switching FET, if it isn't blown? Doesn't need to be anything special anyway, both those part numbers are just bog-standard 8A 600V MOSFETs.

To be perfectly honest, I have no idea! LOL. However in your initial write-up you replaced yours seemingly without hesitation so I assumed it was something that ought to be done on general principle.

How would replacing those resistors diminish the heat they put out, if the ohm value is the same? Ohm's law still applies

That makes sense. I thought a higher watt resistor might not dissipate as much heat since it wouldn't be working as hard. I'm learning this as I go, bear with me. Thanks

If anything, you might wanna re-melt the solder joints of the three regulators, "just in case". Although if they got hot enough to have been a problem, you'd likely see some PCB discoloration in the area.

The regulators and the surrounding PCB all look really good. What would you say about adding heat sinks to these 3 regulators?

You mentioned in the thread that removing the lock nuts for these would be tricky. I haven't taken the main board out yet. Can you elaborate on why they might be diffictult to remove?

Thanks again for your reply. I appreciate the info.

[Khron]

Re: Focusrite Saffire Pro 40 re-cap

Well, i might've been a bit overly-enthusiastic at the time - i've "grown past" that, in the mean time Less free time on my hands, etc.

The locknuts were the "nyloc" type, weren't they? A bit more of a hassle to remove (especially with the long screws used there), but nowhere near impossible.

[SolderFumes]

Re: Focusrite Saffire Pro 40 re-cap

Well, i might've been a bit overly-enthusiastic at the time - i've "grown past" that, in the mean time Less free time on my hands, etc.

The locknuts were the "nyloc" type, weren't they? A bit more of a hassle to remove (especially with the long screws used there), but nowhere near impossible.

That's hilarious! Well, it was 8 years ago when you did the job so you get a pass on that one! Yes, they are nyloc nuts. I thought there might be something under the board that makes them hard to remove.

Sorry to keep bringing it up, but do you think putting heat sinks on the regulators would be worth doing or would it be pointless overkill?

[Khron]

Re: Focusrite Saffire Pro 40 re-cap

I'm at least 70% sure the people at Focusrite designing the thing would've put heatsinks in there, if they deemed it necessary. If the PCB being used as a heatsink isn't discoloured at all, then it's pretty safe to assume the problems you're encountering aren't thermal.

Well, not as far as the silicon is concerned. Electrolytics, on the other hand......