BCD536HP Audio Output Problems

[an39511]

After much $%#^ around trying to narrow this problem down I skimmed right over the people writing about ground loops. Well......That is what is going on. I don't quite understand why it is happening so I will describe the setup. I hope someone can explain why this is happening especially when all equipment is properly grounded.

* Central battery backup (UPS) installed and properly grounded.
* Desktop Computer plugged into the UPS. Grounded, of course.
* Stereo 3.5mm plugged into the LineIn / MIC jack of the computer.
* Connect the other end of the cable to the 536 speaker out jack and sound comes out the left channel. (as expected).

At this point everything seems to be working OK but using a volt meter I am measuring 3.5v between the chassis of the BCD536 to ground. (not good)

If I ground the 536 to the common ground of the computer the sound is cut off. If I connect my outdoor antenna (which is grounded) the sound cuts out.

To narrow the problem down further I unplug the audio cable from the back of the 536 and used a multimeter and measure 2.3v between the tip of the 3.5mm plug to the shield(ground). I also measure 2.3v from the center lead (right channel) to the shield.
NOTE: This voltage measurement is probably normal because I tested this on 2 desktop computers and 1 laptop. All of them show 2.3v to ground. Maybe they use this for powered microphones?

This all boils down to one question; How does the voltage coming out of the audio cable end up charging the chassis of the BCD536?

 

[rgore]

Your UPS and radio are using 2 different grounds?

 

[an39511]

Your UPS and radio are using 2 different grounds?

Short answer is No.
Both devices are connected to the same UPS but the BCD536 has no ground unless you connect a ground wire to the chassis.

 

[RF23]

Short answer is No.
Both devices are connected to the same UPS but the BCD536 has no ground unless you connect a ground wire to the chassis.

A couple more things you might try:

Plug the scanner into a different socket on the UPS, incase the UPS has a bad socket.

Plug the scanner directly into an outlet (by pass the UPS), I have had two UPS units go bad and I don't mean that just stopped working they actually caused some of the gear hooked to the defective UPS to malfunction.

Of course, the 536 may have a ground disconnected or a cold solder joint in it's power supply or the connection between the adapter socket and the internal power supply of the scanner.

 

[UPMan]

Do not ground the shield! The Ext Sp. jack is BTL (both sides driven). If you ground the shield you will short the audio out circuit.

 

[an39511]

A couple more things you might try:

Plug the scanner into a different socket on the UPS, incase the UPS has a bad socket.

Plug the scanner directly into an outlet (by pass the UPS), I have had two UPS units go bad and I don't mean that just stopped working they actually caused some of the gear hooked to the defective UPS to malfunction.

Of course, the 536 may have a ground disconnected or a cold solder joint in it's power supply or the connection between the adapter socket and the internal power supply of the scanner.

Okay. To eliminate the UPS I used a spare power strip plugged into
another outlet and plugged the computer which has a three pronged
plug as well as the BCD536 which only has the 2 pronged wall wart.

Tested for voltage between the chassis of the computer to ground
and voltage from the chassis of the BCD536 to ground.
No voltage flowing on either device.

Connected the 3.5mm stereo audio cable to the computer and the 536.
No voltage flowing from the computer chassis to ground.
3.5v flowing from the chassis of the 536 to ground.

Somehow the voltage supplied on the audio cable is getting to the
chassis of the 536.

 

[an39511]

Do not ground the shield! The Ext Sp. jack is BTL (both sides driven). If you ground the shield you will short the audio out circuit.

I am using a standard factory made cable. Shield of that patch cable is NOT grounded.

The problem is this. If I connect an outside antenna that ground is shorting out the audio. My work around was to make a special audio cable that only connects the center leads and not the shield.

There is something very odd that I don't understand here. How can voltage get from the center lead of the audio cable to the chassis of the 536?

 

[an39511]

Just to add to all of this. The audio problem does not appear when connected to a laptop. Laptops are not grounded. The problem is showing up on desktop PCs which are grounded.

 

[markab]

You may want to visit this link.
http://ncsradio.com/commercial/Data Sheets/Connecting Radios with Bridge Tied Load.pdf

It doesn't pertain specifically to your situation but it does offer a bit of insight into bridge tied load amplifiers. I can't say with certainty but perhaps using an audio isolation transformer would do the trick for you. Good luck.

Mark

 

[JamesO]

Just put a capacitor in series with the shield? This will DC uncouple the shield. Something like a 0.01 or 0.001 MFD ceramic capacitor.

You may also need to do/try something unconventional and use a 2 prong to 3 prong power adapter on the computer? Used to have BIG issues with multiple DirecTV receivers that had 3 prong power cords. Most people just ripped the ground prong out if the cord with pliers, but I kept a bunch of 2-3 prong adapters in my bag of tricks for troubleshooting purposes.

 

[an39511]

You may want to visit this link.
http://ncsradio.com/commercial/Data Sheets/Connecting Radios with Bridge Tied Load.pdf

It doesn't pertain specifically to your situation but it does offer a bit of insight into bridge tied load amplifiers. I can't say with certainty but perhaps using an audio isolation transformer would do the trick for you. Good luck.

Mark

Thank you for this. I think this explains a lot and why my home made patch cable works by only connecting the center lead and the shield floating free. This may explain why other scanners work fine using a stock audio patch cable because they are using a single ended audio as opposed to Bridge Tide Load.

But I am still baffled how the 3.4v voltage is getting from the PC audio cable onto the chassis of the 536. If the audio is indeed isolated how could that happen?

 

[Mike_G_D]

Thank you for this. I think this explains a lot and why my home made patch cable works by only connecting the center lead and the shield floating free. This may explain why other scanners work fine using a stock audio patch cable because they are using a single ended audio as opposed to Bridge Tide Load.

But I am still baffled how the 3.4v voltage is getting from the PC audio cable onto the chassis of the 536. If the audio is indeed isolated how could that happen?

Depending on the design of the differential amp output of the 536, the "BTL" driver described earlier, some form of DC path to ground might be present at either or both of the audio lines. This DC path may be such as to provide high impedance to audio and above ac frequencies but low impedance to DC. Without seeing a full schematic I cannot be certain, of course, so I can only guess. It is possible that the design is such that the DC path to ground is "acceptable" in the internal design as it is intended to be isolated internally. Normally, speakers are passive devices that provide no current source, of course, so they would not be a problem. Also, normally, speakers are usually not designed so as to ground one of their audio leads (the speaker is sort of like a transformer with a diaphragm connected as a secondary). Under these "normal" conditions, the BTL output of the 536 audio will be fine and operate within designed for parameters. However, if you connect the output to a device that is not a speaker and does not use a differential input, such as a basic consumer amplifier or even many if not most PC soundcards, the output could be compromised. Without some form of output protection against such connections and depending, again, on the design of the amp, at best the audio may be limited, distorted, or even cut off completely. At worst, permanent damage to the audio output amp of the 536 may result.

When you also add a source of DC to the line as in the case of many input devices intended to provide DC for "phantom power" for certain microphones, then the problem is further compounded.

I would strongly suggest that the use of an audio isolation transformer be used as described in the first case of the document linked to earlier. That is the best solution.

Otherwise, I would do as the second solution describes in that I would connect the audio cable shield to a ground spot on the radio (chassis ground) and NOT to the speaker "ring" as you might normally be inclined to do with single ended drivers. Then I would connect the center conductor of the cable either to the center conductor of the 536 speaker jack or to its "ring". I would also add an in-line DC filter cap of sufficient size/value to pass audio frequencies with little attenuation but block DC. Something like 10uF would be fine but you might be able to get by with lower values with maybe some attenuation - not too critical but I wouldn't go any lower than 1uF; most of the higher value caps will be polarized so I would "point" the + side to the source of the DC (for example, the PC soundcard) or you could use non-polarized electrolytics or use two polarized ones connected "back-to-back" (+ to + or - to -). This filter cap then would be between the output of the one lead from the 536 audio driver and the input to the PC soundcard or whatever single ended amp is being used. This will fully block any DC from getting into the 536 and will still pass the audio. Understand that this cap is a SERIES cap inline between the output of the 536 audio amp and the input of the device being driven or fed such as a PC soundcard input. Neither lead of the cap should be grounded under any circumstances!!

To summarize:

For 536 audio output from the external speaker jack being connected to anything other than a basic passive speaker I would suggest the following two choices with the first being the best choice:

1) Use an audio isolation transformer with the primary side leads connected to the two "speaker leads" coming from the external speaker jack on the 536 (NOTE THAT NEITHER LEAD SHOULD HAVE ANY CONNECTION TO THE CHASSIS GROUND OF THE RADIO OR THE SHIELD OF THE PROCEEDING CABLE CONNECTED TO THE NON-PASSIVE SPEAKER DEVICE!!!) and the secondary side connected to the leads going to the non-passive non-speaker device or amplified speaker with one lead to the center conductor and the other to the grounded shield of the cable;

2) Use only one lead of the 536 external speaker output (either "tip" or "ring" will do but, for the sake of simplicity and convenience, it is usually best to just use the "tip" lead) and connect that to a DC blocking capacitor of value 1uF or greater with polarized caps having the + lead toward the device being fed/driven using the cable center conductor while the cable's shield is connected to the radio chassis (say with some convenient chassis screw) and NOT to the external speaker output "ring" which must be left floating unless it was used as the source as described earlier instead of the center "tip".

-Mike

 

[an39511]

Mike_G_D : Thank you very much for your detailed description. I have learned a great deal about audio circuits today. I will modify my patch cable accordingly. I hope Uniden adds an entry about this in their manual and/or makes a cable that works properly for those of us that will be streaming online audio from the 536. I am probably lucky I didn't burn something out with all my experimenting. I wouldn't have bothered to use the rear audio jack if there wasn't such a problem with all the noise on the headphone jack.

 

[cpetraglia]

My question is why are you using the speaker out which is designed for an 8 ohm load and has too much power for the input of the PC card? I use the head phone out through a GLI with no problems. Driving a high ohm low voltage input with speaker power is sure to create problems and could damage your sound card.

 

[wm8s]

Do not ground the shield! The Ext Sp. jack is BTL (both sides driven). If you ground the shield you will short the audio out circuit.

Is this why we cannot use an amplified speaker? Is that common in the industry for these kind of radios' external speaker connections? I had hoped to hook this to the radio input on a vehicle's PA, but I guess that's out of the question, yes?

 

[K9DAK]

Mike_G_D : Thank you very much for your detailed description. I have learned a great deal about audio circuits today.

Wow, this is something that we concert audio techs deal with all the time . . . balanced (BTL as referenced earlier) vs. unbalanced (single ended) audio connections from mics, guitars, etc., to our mixing boards. Some boards are designed so that you can plug a TRS (tip-ring-sleeve) connector half-way into the jack to isolate the ground.

I would never have expected this issue in consumer-grade equipment. One should be able to take an off-the-shelf cable and plug it into any speaker, amplified speaker, or computer and just have it work without adding transformers or making special cables.

Bad design choice on Uniden's part if you ask me.

 

[an39511]

My question is why are you using the speaker out which is designed for an 8 ohm load and has too much power for the input of the PC card? I use the head phone out through a GLI with no problems. Driving a high ohm low voltage input with speaker power is sure to create problems and could damage your sound card.

The sound from the rear jack does not overpower the input on the PC. If you scroll up you will see an explanation on how I made the cable. I just didn't understand why it worked but now I understand about how the amplifier is designed.

The reason I didn't use the headphone jack is due to the audio problems discussed in another thread.

 

[an39511]

Wow, this is something that we concert audio techs deal with all the time . . . balanced (BTL as referenced earlier) vs. unbalanced (single ended) audio connections from mics, guitars, etc., to our mixing boards. Some boards are designed so that you can plug a TRS (tip-ring-sleeve) connector half-way into the jack to isolate the ground.

I would never have expected this issue in consumer-grade equipment. One should be able to take an off-the-shelf cable and plug it into any speaker, amplified speaker, or computer and just have it work without adding transformers or making special cables.

Bad design choice on Uniden's part if you ask me.

I don't know if I would call it a bad design but I am guessing it will cause a lot of confusion for many people that know nothing about BTL. I would expect there are plenty of people that are planning to stream audio online and if they are using a PC to do it they are in for a surprise. With the audio problems on the headphone jack they rear jack is where they are going to try and hookup.

 

[Boatanchor]

Of course, if Uniden had retained the audio line output, none of these problems would be occurring in the first place!

All professional and semi-professional receivers have audio line outputs. It does seem a bit ridiculous to eliminate the line output on your 'flagship' receiver product for the sake of saving a couple of bucks.

 

[Mike_G_D]

Differential outputs on audio amplifiers are actually a "good thing" at least from a purely theoretical standpoint. In addition to the power efficiency benefit described in the document linked to earlier (for the bridged tied load, "BTL", approach) you also gain noise immunity. Common mode noise is isolated and significantly reduced relative to a single ended design with one lead grounded.

That having been said, one does need to consider overall system design within the expected common user environments.

Had I, for example, been given the task of designing the external speaker audio circuit of the 536, I would have done so with the expectation that many users would attempt to connect the business end to something other than a passive ground isolated speaker, especially considering that a line level record output which had been standard issue on previous models was now being omitted in the final design. Given that situation, I either would have stayed with the common single ended approach or employed a differential output with sufficient protection against common mode single ended termination issues including DC isolation and zero impedance to ground protection. The latter would have added to the component count and overall cost but would have been a very prudent approach if differential output was a general mandate.

I can only guess that differential output audio was an expedient and efficient solution for, maybe, two reasons.

I think that it has been mentioned that the 536 uses a class D audio amp driver; this approach provides wonderful power efficiency for a given supply voltage with very good audio reproduction but does add complexity which may have extra sensitivity to common mode issues including induced line noise. I'm not sure as my class D amp knowledge is too limited currently and I would have to study it further. But these beasts are way more complex than a simple class A amp which is very pure but very inefficient, power-wise. Differential outputs may be the most common form of such audio amps. Or they may be necessary to protect against unstable operation (such as oscillation) in the presence of high noise levels as might be encountered on heavily populated and dense RF/IF/AF radio receivers with RF, IF and digital noise sources present en mass.

It may also have been expected that many users would employ very long leads between the radio and the external speaker in mobile electromagnetically noisy environments in which case a differential output would make considerable sense; I suspect that is why many professional LMR radios use this approach.

Still, given the hobbyist user base expected and the likelihood of many of said users employing non-passive non-speaker terminations for the audio output, again, I would have insisted on adequate protection and, if at all possible, the use of an approach that allowed single ended OR differential terminations and included adequate DC blocking protection.

I do not own and cannot afford to own a 536 and have not read the manual and specifications in detail but I was under the, perhaps mistaken, impression that the digital interface, USB, if I understand correctly, does contain audio information. If so, then perhaps the "powers that be" expected the users to use that means for audio input to computer devices for the purpose of streaming, remote operation, etc. If true, this would also help to explain the omission of a line level "record output" jack as was present on previous models.

-Mike