• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

How To Build Audio Electronics In America

Jim Shaw

Addicted to Fun and Learning
Forum Donor
Joined
Mar 16, 2021
Messages
501
Likes
921
Location
Northeastern Ohio, USA, in the woods
I just saw this YT video and found it so interesting that I thought to reference it here. Highly recommended background info.

I give John Darko on YT mixed reviews at best, but his videographer definitely knows what she is about. In this video, she travels with Jason Stoddard of Schiit Audio (California here, and Texas). The resulting exposition of how audio equipment gets built is really the best I've ever seen. Schiit is a good example because they are vertically integrated (making much of what they sell), American (English is their first language), and very open to showing 'how the sausage is made.'

Aside: I've been an electrical engineer in the avionics business for over three decades (now retired). I have never seen a better video coverage of how audio electronics are made.
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,753
Location
Santa Fe, NM
Yes, quite good. I am a bit surprised that in final test they do complete electrical testing only on a small sampling of units. All our amps are tested before burn-in and retested after burn-in on an Audio Precision 5xx analyzer, and defects sometimes do show up during these tests. Also in some units, the firmware has built-in self diagnosis features which complements the final testing process and also allows the end user to read fault codes stored in the amplifier in the event of something going wrong. This can save the customer having to send complete (and heavy) amplifiers back to the factory.
 

clearnfc

Active Member
Joined
Dec 8, 2021
Messages
236
Likes
68
When you look at the production, you will realise that there isnt much value for it to be made in USA anymore...

Components come from everywhere around the world now (usa dont even make most of capacitors or mostfet/bipolar transistors, mostly japanese) and they are just assembled by machines or by hand.

If you look at the factories in china today, they can do it just as well, if not better for a fraction of the cost.
 

restorer-john

Master Contributor
Joined
Mar 1, 2018
Messages
9,961
Likes
28,421
Location
Gold Coast, Queensland, Australia
I am a bit surprised that in final test they do complete electrical testing only on a small sampling of units.

Different price point for your amps vs Schiit's single board-in-a-can wouldn't you say?
 

kongwee

Addicted to Fun and Learning
Joined
Jan 22, 2022
Messages
733
Likes
190
When you look at the production, you will realise that there isnt much value for it to be made in USA anymore...

Components come from everywhere around the world now (usa dont even make most of capacitors or mostfet/bipolar transistors, mostly japanese) and they are just assembled by machines or by hand.

If you look at the factories in china today, they can do it just as well, if not better for a fraction of the cost.
Some of its' press machines comes from China.
 

MarcosCh

Major Contributor
Joined
Apr 10, 2021
Messages
1,455
Likes
1,095
Well i think the guy makes it clear, that the objective is to contribute to the local economy, and that this can add value to some customers, and others don't care.
Agree that in general the videos in darko channel are well made and that makes them easy to watch regardless of the content. A lot to learn for other youtubets
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,753
Location
Santa Fe, NM
Different price point for your amps vs Schiit's single board-in-a-can wouldn't you say?
I haven't compared prices between the amps, so I couldn't say. Still, it only takes a short amount of time to run an automated Audio Precision test suite on a product, so I'm still baffled that complete testing isn't done.
 

Gringoaudio1

Active Member
Forum Donor
Joined
Sep 11, 2019
Messages
211
Likes
261
Location
Calgary Alberta Canada
I haven't compared prices between the amps, so I couldn't say. Still, it only takes a short amount of time to run an automated Audio Precision test suite on a product, so I'm still baffled that complete testing isn't done.
Pretty standard practice in manufacturing to just test a sample of the production. Test everything for the first production run to find the number of flaws you find to generate an estimation of how many returns you’ll get. Then you compare the cost of testing and slowing down the production line versus the estimated cost of warranty work. That comparison determines how many you test for subsequent production runs.
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,753
Location
Santa Fe, NM
Pretty standard practice in manufacturing to just test a sample of the production. Test everything for the first production run to find the number of flaws you find to generate an estimation of how many returns you’ll get. Then you compare the cost of testing and slowing down the production line versus the estimated cost of warranty work. That comparison determines how many you test for subsequent production runs.
Except when your product is pushing 100 pounds and you really don't want to have to ship it back to the factory and back to the customer.

There are a lot of errors which can creep into products during assembly; complete testing finds these. I'm not terribly impressed with 'standard practice in manufacturing' and I'm sure that if your reputation and paycheck were relying on spot testing, you might just want to be a bit more comprehensive. :facepalm:
 

Gringoaudio1

Active Member
Forum Donor
Joined
Sep 11, 2019
Messages
211
Likes
261
Location
Calgary Alberta Canada
True enough with big products. The cost of an RMA would include shipping. Schitt products are tiny. But to avoid returns you invest more in your production line. And again the cost of quality control in the line is weighed against the costs of Returns. I was in China at a vendor: Haier. They were assembling and testing remotes. After they were snapped together the person doing that would whack the remote against the edge of the counter. Some were being whacked three times and somewhere being whacked six times before being tested. All were tested. Probably to more or less resolution depending on the product. I asked our tour guide why the different number of whacks. She said, “Client pay for more”.
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,753
Location
Santa Fe, NM
True enough with big products. The cost of an RMA would include shipping. Schitt products are tiny. But to avoid returns you invest more in your production line. And again the cost of quality control in the line is weighed against the costs of Returns.
Uh, we're primarily concerned with providing the customer a product which will last basically forever and our reputation for durability, build and reliability is well established. It didn't get that way by letting the bean counters run wild with cutting costs. ;)
 

Gringoaudio1

Active Member
Forum Donor
Joined
Sep 11, 2019
Messages
211
Likes
261
Location
Calgary Alberta Canada
Making a product that lasts forever is not the way to make money in ‘mass manufacturing’. Statistics are used to generate estimated lifetime and statistics are used to calculate all the costs required to make it so and it is definitely well under 10 years so that the customer will buy another one. There certainly are boutique brands of audio that can get passed down to your children. But you pay for that upfront with a high cost of entry.
And bean counters do rule the world. My ex was one!
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,753
Location
Santa Fe, NM
Making a product that lasts forever is not the way to make money in ‘mass manufacturing’. Statistics are used to generate estimated lifetime and statistics are used to calculate all the costs required to make it so and it is definitely well under 10 years so that the customer will buy another one. There certainly are boutique brands of audio that can get passed down to your children. But you pay for that upfront with a high cost of entry.
And bean counters do rule the world. My ex was one!
Fine whatever.
 

JeffS7444

Major Contributor
Forum Donor
Joined
Jul 21, 2019
Messages
1,962
Likes
2,869
Good video, and excellent publicity for Schitt: Customers do love a story to go with their luxury purchases.

Brand USA still has perceived value as does "Made in Germany", but the latter label allows for a certain percentage (apparently quite a lot) of non-German content and labor.
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,753
Location
Santa Fe, NM
Good video, and excellent publicity for Schitt: Customers do love a story to go with their luxury purchases.

Brand USA still has perceived value as does "Made in Germany", but the latter label allows for a certain percentage (apparently quite a lot) of non-German content and labor.
A lot of the made in America difference is the mindset of the people who are building the product and the personal investment they have in it. I'm sure, say, MacIntosh has a similar mindset.
 
OP
Jim Shaw

Jim Shaw

Addicted to Fun and Learning
Forum Donor
Joined
Mar 16, 2021
Messages
501
Likes
921
Location
Northeastern Ohio, USA, in the woods
Pretty standard practice in manufacturing to just test a sample of the production. Test everything for the first production run to find the number of flaws you find to generate an estimation of how many returns you’ll get. Then you compare the cost of testing and slowing down the production line versus the estimated cost of warranty work. That comparison determines how many you test for subsequent production runs.
Yes indeed. There's an entire science, mathematics, and professional discipline devoted to sampling and testing. There are some good reasons for 100% complete testing and burn-in. There are also bad reasons. The statistics guys can help a lot in deciding what to test and how completely to test. Test regimens can be anything from looking at the pilot light to a few test point readings to a complete in-situ testing regimen like we do in aircraft avionics. Simulating in situ testing is a surprisingly costly process, often including test equipment costs up into the millions. Real, degreed, practicing engineers have had at least a modicum of the testing discipline, and they should know where to get the numbers they need. It's why tinkerers often have problems.

Of course, you would not go that far in testing with a device whose biggest result of failure would be silence, a buzz, or a wisp of smoke out the vents. Absolute zero failures is incredibly expensive.

Um... "In all things moderation..." is how testing works best. The Devil is in the details. And "experience is what you learn that you wish you didn't."

.........
PS: Modern electronics often have "Built-In_Test" capability. The chip-based processor that monitors bias, current, and voltage for proper and safe operation often has the protective functions overlayed with software that perform a startup test before the device is fully functional. Such "BIT" capabilities really help to provide fast tests -- including a part of final quality control tests.
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
12,199
Likes
14,290
Location
Riverview FL
Not seen the Chopsticks Method of soldering before, at 4:37.

That must have come from Asia.

1649714599556.png
 
Top Bottom