HDMI connections can be a source of non resolvable Ground Loops

[TLS Guy]

Since getting a new Direct TV DVR (Gennie) I have had a ground loop issue. This is a ground loop related to HDMI in the DVR and NOT resolvable.

I had no issues with my first Direct TV DVR. When it failed the replacement buzzed like crazy, with no solution. So Direct TV provided their Gennie unit a couple of years ago. This is better but still has a slight buzz when connected to the pre/pro via HDMI. You can just hear it at the seated position it does not bother TV watching, but does bother music listening. You can just hear it in quiet passages. The hum is the same whether the Direct TV unit is on or off. It is present if the HDMI cable from the unit is plugged into the pre/pro. Disconnect the cable and everything is dead quiet.

Now the FM antenna grounds and the dish grounds are properly bonded. Disconnecting the dish cable to the DVR stops the buzz. But if the ground of that cable is connected to the system there is no buzz. If the DVR is connected by HDMI to the pre/pro if the DVR is grounded in any way the buzz returns. I should add there is no buzz if the pre/pro is off.

If you connect the outside ground of the HDMI plug to the AVR with alligator clips there is no buzz. The HDMI connector has to be put into an HDMI socket for this to return. This was the same situation for two different pre/pro models.

I have solved the situation by putting a small link lead in the back of the DVR and only connect the DVR to the pre/pro while watching TV. At all other times it is disconnected. This works but is not handy, especially for my wife.

So this proves the Direct TV DVR has an internal ground loop between chassis and he HDMI connection.

This is the pin out of HDMI.

Type A HDMI connector pinout
Pin Signal Description
1 TMDS Data2+
2 TMDS Data2 Shield
3 TMDS Data2-
4 TMDS Data1+
5 TMDS Data1 Shield
6 TMDS Data1-
7 TMDS Data0+
8 TMDS Data0 Shield
9 TMDS Data0-
10 TMDS Clock+
11 TMDS Clock Shield
12 TMDS Clock-
13 CEC control
14 Reserved/HEC Data− N.C. on device
15 SCL DDC clock
16 SDA DDC data
17 DDC/HEC/CEC Ground
18 +5 V Power power EDID/DDC
19 Hot Plug Detect/HEC Data+

Now you will note that there are 5 ground pins, in addition to shield connected to the outside of the HDMI plug.

These other grounds can not be true grounds, and here in lies the problem.

Yesterday I had a long discussion with Blue Jeans Cable about this. To my surprise they were well aware of the issue. According to them this affects quite a few HDMI connections.

I phoned them to ask if they could make me a custom cable to bond all these grounds. They told me they had already thought of that and have done it. However all units connected by a cable modified in that way shut down. They have no solution.

So thinking about this, some or all of these pin grounds are not true grounds, but have a control function. In other words there must be some sort of control or sensing circuit between the so called ground cables and true ground. Now this Direct Genie DVR is the only unit out of six that has any sort of buzz when plugged into the pre/pro. So it means that my high quality units have very low resistance sensing circuits on those so called ground pins. Direct TV being corporate must have cheap components with excessive resistance to ground setting up the internal ground loop.

This is yet another reason why HDMI has real downsides to audio quality.

In my view we should move to an optical connection with the same functionality as HDMI. That would be the ideal solution.

 

[Irvrobinson]

Well, this might explain my unresolvable hum in my HT system. I have tried to avoid thinking too much about HDMI, because it is not a specification to my liking, to say the least. IMO, the right answer is HDBaseT:

https://hdbaset.org/

But it'll take a while to get broad adoption. The right companies are involved though, so there's hope.

 

[TLS Guy]

Well, this might explain my unresolvable hum in my HT system. I have tried to avoid thinking too much about HDMI, because it is not a specification to my liking, to say the least. IMO, the right answer is HDBaseT:

https://hdbaset.org/

But it'll take a while to get broad adoption. The right companies are involved though, so there's hope.

The trouble with HDBaseT is that it is not the all in one solution that the promoters claim. The fact is that it is complimentary to HDMI and not a replacement. It does not address DRM so the big shot program providers will not go for it. It requires HDMI modules at both ends for DRM protected material. So it can transmit very high def audio and video over long distances, plus control functions and up to 100 watts of power.

I don't see it having a place in most home HT applications at present.

The next issue will be the howls related to a obsolescence if there is a move to abolish HDMI on new units. Unless we are very lucky we are stuck with that ill thought out and conceived HDMI/HDCP system for the foreseeable future.

 

[Irvrobinson]

The trouble with HDBaseT is that it is not the all in one solution that the promoters claim. The fact is that it is complimentary to HDMI and not a replacement. It does not address DRM so the big shot program providers will not go for it. It requires HDMI modules at both ends for DRM protected material. So it can transmit very high def audio and video over long distances, plus control functions and up to 100 watts of power.

I don't see it having a place in most home HT applications at present.

The next issue will be the howls related to a obsolescence if there is a move to abolish HDMI on new units. Unless we are very lucky we are stuck with that ill thought out and conceived HDMI/HDCP system for the foreseeable future.

I think you're missing my point. HDBaseT is protocol agnostic. It can encapsulate a protocol in 802.3, so for the content industry it can transmit HDMI-compatible protocol without all of the electrical stupidity of HDMI cabling. The problem in your original post goes away. I was only referring to the electrical issues caused by HDMI's stupid electrical cabling scheme. While HDMI as a protocol may take a long time to go away, hiding it within HDBaseT, IMO, would be a first good step.

 

[TLS Guy]

I think you're missing my point. HDBaseT is protocol agnostic. It can encapsulate a protocol in 802.3, so for the content industry it can transmit HDMI-compatible protocol without all of the electrical stupidity of HDMI cabling. The problem in your original post goes away. I was only referring to the electrical issues caused by HDMI's stupid electrical cabling scheme. While HDMI as a protocol may take a long time to go away, hiding it within HDBaseT, IMO, would be a first good step.

Yes, I agree it would have to sit side by side. However if there will need to be electronics either end to make it plug into HDMI ports. I agree though it is a much better transmission system.

Ideally though we should ditch HDMI entirely.

 

[Irvrobinson]

Ideally though we should ditch HDMI entirely.

On this point we completely agree, but the video and music content industry is so protective of control over distribution, and so adverse to technological progress in favor of consumers, that I suspect protocol encapsulation may actually extend HDMI's life, not shorten it.

 

[TLS Guy]

On this point we completely agree, but the video and music content industry is so protective of control over distribution, and so adverse to technological progress in favor of consumers, that I suspect protocol encapsulation may actually extend HDMI's life, not shorten it.

Yes, it will. I agree completely. That is another of HDMIs problems.

 

[slipperybidness]

Yes, it will. I agree completely. That is another of HDMIs problems.

@TLS Guy
@Irvrobinson
The big problem with DRM is that it really only hurts the honest end-users. All of this effort from the studios, but it still gets defeated by the people that they try to lock out, then the honest users are the only users that suffer.

 

[Irvrobinson]

The big problem with DRM is that it really only hurts the honest end-users. All of this effort from the studios, but it still gets defeated by the people that they try to lock out, then the honest users are the only users that suffer.

The efficacy of DRM is an entirely separate issue. The music industry obviously thinks they need it, and their stance on DRM is quite simple. They own the content, they legally have control over what you can do with it, and if you don't like DRM then don't buy their content, buy someone else's. We have no legal right to anyone's artistic creations, and they have no obligation to sell you movies, videos, and music at all. And all of this posturing is defended by US and European laws that come with big fines for violations. (In the US, six-figure fines.) I view DRM as permanent annoyance that isn't going away in my lifetime.

 

[Bucknekked]

@TLS Guy
@Irvrobinson
The big problem with DRM is that it really only hurts the honest end-users. All of this effort from the studios, but it still gets defeated by the people that they try to lock out, then the honest users are the only users that suffer.

I agree. the folks that DRM puts in the most difficult situations are the law abiding folks who want to buy their content (not steal it) and pay for it. After that, the shackles are on the the hurdles start. I don't want to hijack this very fine thread with a side pet peeve. I mean, how many times do I have to buy the White Album? sheesh.