Juma's Head Amp

I was thinking, how about a simple UPC1237 based with a DPDT relay protect with a DC detect threshold of +/- 1.5 to 2 V DC. PCB about 40x40 mm. Any comments?
Atleast, phones should be able to survive 2V dc.
reg
Prasi

I like the idea - and it is certainly needed. I think the max voltage threshold may be lower for some headphones - in the 500mV range even. Maybe have adjusted low and high threshold? High impedance phones new dhigher voltage drive while low impedance phones are lower voltage.

Here is a nice sealed relay with gold contacts.
https://www.omron.com/ecb/products/pry/111/g6k.html
 
I like the idea - and it is certainly needed. I think the max voltage threshold may be lower for some headphones - in the 500mV range even. Maybe have adjusted low and high threshold? High impedance phones new dhigher voltage drive while low impedance phones are lower voltage.

Here is a nice sealed relay with gold contacts.
https://www.omron.com/ecb/products/pry/111/g6k.html

thats a nice relay...

I do not know if upc1237 can be made to operate below +1.3VDC threshold, based on my rough calculations.

UPC1237 pdf, UPC1237 description, UPC1237 datasheets, UPC1237 view ::: ALLDATASHEET :::
 
The AC Volt measurement function of the Fluke 101 is only specified from 40Hz to 500Hz, which means noise in the remaining audio range (501 - 22kHz) are most likely not covered.
Although the DMM are not "flat" across the audio range they are a lot better than nothing.
I found that my very old DMM circa 1986, works out to 50kHz around which it plummets.
There is noticable rise in response above 5kHz to 10kHz which will exaggerate noise before it begins to fall again above 20kHz to 30kHz.
 
I have found a correlation between my Fluke 101 and subjectively how my own hearing is. It seems to cover what is audible to my ears very well. So 0.0mV on the Fluke is dead as a nail silent. Meaning cannot tell device is on or connected with ear pressed to the membrane (or headphones pushed against ear).
 
I have found a correlation between my Fluke 101 and subjectively how my own hearing is. It seems to cover what is audible to my ears very well. So 0.0mV on the Fluke is dead as a nail silent. Meaning cannot tell device is on or connected with ear pressed to the membrane (or headphones pushed against ear).
This depends very much on the sensitivity of the speaker and how close one is to that speaker. Desktop monitoring speakers, cf. PA speakers 10m away in a big theatre.
0.1mVac at the input terminals of a 86dB/W @ 1m speaker will be quite different to the same noise voltage applied to a 96dB/W @ 1m speaker and very different again to that from a 100dB/mW @ the ear headphone. 100dB/mW @ the ear is equivalent to 130dB/W @ the ear.
 
it's not whether different speakers make the noise audible.
It's about your assertion that 0.1mVac resolution of your DMM is a good guide to inaudibility.
You probably need a (accurate) resolution down at 0.01mVac, or maybe lower, to check for audibility.

A +20dB, or preferably a +40dB Low Noise Amplifier (LNA) +0.1mVac resolution would help a lot in getting any correlation to audibility for a wide range of amplifiers. Particularly the Follower type that are both low in noise and low in distortion.

Some while back I bought Frex's PCBs to create a test set but have not assembled the quite complex circuits, yet.
Yesterday I received the Euvl/Groner LNA that will let me "see" more of these very low level problems on my scope.
 
Last edited:
The specified offset voltage detection levels at pin2 for an invert @ pin6 is: +0.62Vdc & -0.17Vdc ±tolerance.
even if we make RA/RC=0 by making RC=infinite, still the +Vth= 2*0.62=+1.24VDC
page 4 of datasheet
 

Attachments

  • upc1237+vth.png
    upc1237+vth.png
    19.2 KB · Views: 303
Hi,

The issue with UPC1237 is that it works properly from 25-60Vdc (see spec. sheet).

Most of the kit found on e-bay use this IC at about 15Vdc, it can maybe work but not for long and not in a reliable way. This has been my experience.

Is there a similar IC that can work at a lower rail voltage. Having to use 25Vdc and up on a Headphone Amp. is pretty rare and not convenient unless you use a separate dc supply.

If adding a PSU is an issue, maybe we could create a simple voltage doubler to get to the right UPC1237 required operating voltage...

BR,
Eric
 
Last edited:
This is confusing with their own spec. sheet showing the recommended operating voltage....lol (I'll check if my el-cheapo speaker protection are wired with 3.4V directly on pin8. Thanks for the tip.)

You would then need to add a regulator, maybe a LM317 to get to 3.4Vdc
 
Hi,

The issue with UPC1237 is that it works properly from 25-60Vdc (see spec. sheet).

Most of the kit found on e-bay use this IC at about 15Vdc, it can maybe work but not for long and not in a reliable way. This has been my experience.

Is there a similar IC that can work at a lower rail voltage. Having to use 25Vdc and up on a Headphone Amp. is pretty rare and not convenient unless you use a separate dc supply.

If adding a PSU is an issue, maybe we could create a simple voltage doubler to get to the right UPC1237 required operating voltage...

BR,
Eric

These are real handy for getting a higher voltage from your PSU.

DC-DC Adjustable Boost Buck Step Up Converter Voltage Module Power Regulator | eBay
 
Yes that is all that is needed. So one PCB with integrated PSU and now DC protection.... I wonder where those good ideas came from but keep on the good work 🙂

Now we need a guy that will tell to integrate the UPC1237 and relay on the one board design...
 
Last edited:
Yes that is all that is needed. So one PCB with integrated PSU and now DC protection.... I wonder where those good ideas came from but keep on the good work 🙂
you know where the ideas pour out, thanks, i will keep on doing the good work
Now we need a guy that will tell to integrate the UPC1237 and relay on the one board design...
we already have one here named jean-paul😉 for your kind info🙂
 
With relay there will be signal loss. I have the upc based soft-start/speaker protect for an diy amp, when I am not using the relay vs using the relay i can tell the difference.

I would not give more than 0.5v to my phones.

Sent from my A0001 using Tapatalk
Unless it uses a high quality relay , you will always hear a different. even the best ha protects i have seen here use a relay. another option is to use a crowbar protection, the amp will be in la-la land....not sure if a crowbar protect like the apex one can be designed for an HA. Mr. Mile can guide us here...