Using old phones with Ooma Telo - why doesn't this work?

If you still have that POTS line, check the off-hook (open-circuit) voltage, and then check the on-hook current through the phone.

The Ooma hardware is probably either supplying insufficient voltage, or has too high a DC resistance, or both.

--
Tim Wescott 
Control system and signal processing consulting 
www.wescottdesign.com

Some things to try...

Reverse the polarity to your phone - take the cover off and swap the red and green wires in the line cord. The Touch-Tone pad may stop working when you do this. If that happens, use a modular "splitter" to plug in a modified and unmodified phone. Pick up the unmodified phone, dial, then pick up the modified phone and hang up the unmodified one. Leave a message to see if the audio is any better.

Also, if the yellow and black wires from the line cord are connected to anything inside the phone, note where they are connected, and disconnect them, to see if that makes a difference. If that helps, tape off the yellow and black wires from the line cord and run the phone that way.

You can measure the DC voltage across the red and green wires in the line cord. With the phone hung up, it should be 48 V DC, but this isn't super critical. When you pick up the phone, the voltage should drop. If it goes all the way down to 1 or 2 volts you will have trouble; if it stays near 48 volts you may also have trouble.

This is less likely, since it works OK on a real land line, but: unscrew the transmitter cover, take out the carbon microphone capsule, hold it up to your ear, and shake it. You should hear a "sshh-sshh" swishing noise from the carbon granules shaking around. If not, bang the capsule on the table a couple of times. Also make sure the contacts on the bottom of the capsule, and the springs in the handset that touch them, are not covered with gunk.

This is just the ring generator inside the Telo. The circuit that supplies "talk battery" (the DC voltage to run the phone) will be different and is maybe not up to the job - measuring the DC voltage at the phone will tell you that.

Matt Roberds

Take the handset and bang the microphone on a table 1 or 10 times. It is likely the carbon granules are clumped together in the mic element. Mechanical shock will loosen them up and make the mic work better.

Yep. My guess as well. Not enough loop current to run a carbon mike properly. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| Phoenix, Arizona  85048    Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

My first job was at a company that did phone stuff. IIRC, an old-style phone works OK with a 24V battery, but I can't remember what we used for a loading resistor. It's less than 25 years ago -- I think I'm going senile!

--
My liberal friends think I'm a conservative kook. 
My conservative friends think I'm a liberal kook. 
Why am I not happy that they have found common ground? 

Tim Wescott, Communications, Control, Circuits & Software 
http://www.wescottdesign.com

I found this tidbit...

"The normal voltage of a CO line is about 48-50 VDC, and the Loop Current of the line should be between 23ma and 27ma range, but in some instances that may not be true. Most phone systems work fine with Loop Current in the 23ma to 35ma range, to avoid any problems it is recommended to be 23ma to 27ma range." ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| Phoenix, Arizona  85048    Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

Sound too much like Obama to me :-)

When we moved here I found that ours is only around 20mA and some speakerphone don't work reliably with that. Missy Bell said that's ok, or at least that they ain't gonna do nothing about it. So I had to get a speakerphone that works with this. Panasonic Easa-Phone did the trick, in case any readers run into a similar situation.

--
Regards, Joerg 

http://www.analogconsultants.com/

Going? No comment!!! ;-)

I appreciate all the great suggestions. I've done some testing:

The on-hook voltage of the Telo is 46.2V.

Telo off-hook voltage using the old analog phone continuously fluctuates between 6 and 8V, and that appears to vary with the audio input. However, if I remove the carbon micropohone disk, it's 9.42V, and rock solid.

Telo off-hook voltage using the newer electronic phone is rock solid at 5.71V no matter what the audio input is.

So then I decided to try all that on my land line. On-hook voltage is 49.9V. Off-hook, the old phone fluctuates from 7 to 9V, and is solid at 10.42V with the microphone removed. The electronic phone is 6.88V off-hook, solid.

I'm not clear why the Ma Bell off-hook voltage would be higher than the Telo in all cases. But of course we don't know what the Telo is putting out. And it seems the resistance of the mile of 22-gauge twisted pair copper on the Ma Bell side should have something to do with the difference, but every time I try to figure that out, I conclude I'm thinking about it backwards.

I made sure polarity was the same in all cases. I didn't do any current measurements.

Doing some further reading, it seems the carbon mic is actually a variable current device. Of course the current changes get converted to voltage changes somewhere in the process, but I think this technology is just incompatible with the Telo. The fact that the voltage on the line is so eratic probably doesn't help the Telo extract the actual audio signal from it, and it ends up doing a poor job. Ma Bell retains the ability to handle carbon mic input because of people like me, but I think the Telo people probably just concluded it wasn't worth the trouble. Actually, I pretty much agree with that.

In any case, and whatever the explanation, it seems the answer is to get a couple newer electronic phones, and I've ordered one from Amazon to try out.

Carbon microphones are variable resistance devices: the sound pressure compresses and rarefies the carbon granules in response to sound pressure, which makes the resistance go down or up. That changes the current through the phone into a voltage.

Apparently the very earliest telco amplifiers were a speaker acoustically coupled to a carbon mic, and the assembly actually achieved power amplification (with enough drawbacks that Ma Bell was one of the earliest adopters of vacuum tube technology).

--
Tim Wescott 
Control system and signal processing consulting 
www.wescottdesign.com

This brings back fond nightmares. I ran a similar measurement on my POTS line. Incidentally, I'm 9500ft from the CO.

On hook is about 50.0 volts[1]. I tried various modern instruments, all of which dropped the line voltage to about 9.2 to 9.5 volts. My antique Northern Electric rotary dial butt-in should be a fair simulation of an antique phone. It dropped the line voltage to about

9.3 volts. I confirmed the measurements with a Triplett Model 5 Loop Tester and an Ameritec Model AM-44 VF tester (just to see if they still were working). Loop current with a 600 load was 31.3ma. 5.71v seem far too low, even for your new electronic phone. It should be more like 9 - 10VDC. Just to be sure, I tried about 8 assorted instruments, all of which showed 9 to 10V off hook. Low DC and probably low loop current, is a sufficient excuse to justify a (hopefully free) service call from your local monopoly.

I don't have an Ooma box handy, but I have a pile of Linksys PAP2T-NA awaiting installation. 49V on hook, 9.5V off hook, 19.5ma and 20.3ma loop current (L1/L2), using the Triplett tester at 600 ohms.

I'm not sure what's wrong at your end, but these number may help. As was suggested, try measuring the loop current. Anything less than

20ma will indicate a problem.

[1] Measured with my Fluke 77 DVM, after replacing the leaking 9V battery. My Triplett had two leaky 9V batteries and my Ameritec had 4 dead AA batteries that were trying to leak. Time for a general battery inspection and purge before electrolyte rot kills all my test equipment.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

On a sunny day (Sat, 23 Feb 2013 13:58:26 -0800) it happened Jeff Liebermann wrote in :

Boeing may have some cheap Li-ion for you... :-)

This is maybe a little low, but it doesn't actually matter that much. The telephone pretty much doesn't use DC voltage in this state; it is allowed to leak a tiny amount of current, but that mount is not enough to do anything useful.

If the room is quiet, the voltage should be fairly steady, but there will be small fluctuations - the microphone is doing its job of sending audio down the line.

On the other hand, if the Telo really is being loaded down by the carbon microphone, its voltage output may be changing at the same time the microphone is also trying to change the voltage, which may be the cause of the bad sound.

The Telo box is probably powered by an AC adapter that fits in your hand and weighs a couple of ounces.

At your friendly local Ma Bell exchange (the brick building that looks like it can survive a tornado, earthquake, and Godzilla simultaneously

- because it can), in the basement, there are 24 two-volt cells. Each cell is about half the size of the dishwasher in your kitchen, and they are bolted together with copper bars that are thicker than your thumb. Ma replaces (or used to replace) the cells every 20 years or so whether they need it or not.

That's why. :)

(Okay, in these latter days, your battery and dialtone is likely to be coming from those cabinets that are popping up through neighborhoods everywhere. These have four 12 V absorbed glass mat or gel-cell batteries wired in series; each battery is about the size of one or two car batteries.)

The resistance of the mile of wire does have an effect. If you carried your old phone to the exchange and plugged it in right there, you would see (say) 48 V on-hook and 47 V off hook. Put a mile of wire in between you and and the exchange, and you get the drop to about 10 V off hook that you see.

The important thing is that the batteries at Ma can sustain that 10 V voltage at your phone very well. When you talk, the resistance of the microphone decreases that voltage a little bit and then lets it come back up, but it always comes back up to a steady 10 V.

The Telo box is dropping to 8 V, and I suspect it's hanging on "by the skin of its teeth" to do that. When you start talking, the voltage put out by the Telo box is probably sagging down, *in addition* to the voltage decrease caused by the microphone.

The Telo box doesn't have the mile of wire to work against; I suspect it is either very current-limited, which would make the voltage drop when the phone is picked up happen "automatically", or it is programmed to limit the current or drop the voltage when you go off hook, to simulate a real Ma Bell line.

I agree with that for a short-term answer. The long-term answer is to complain to Ooma that their box can't operate a standard telephone. :) (There is a small but nonzero possiblity that this can be fixed with some kind of settings change or software upgrade. However, it's more likely that the Telo box needs a better design for its internal power supply.)

I had an interesting chat with the guy that was installing the new whiz- bang VoIP PBX at a previous workplace. I asked if it could support analog phones - yes. How about rotary dial? He laughed and said, "Well, *now* it can." Apparently the first units could only do analog Touch-Tone. This worked great until they got installed in churches and elementary schools. It seems that those places tended to still have the Western Electric rotary-dial phone from 195x or 196x still on the wall... in the kitchen.

The first attempted fix was usually to replace that phone with a whiz- bang electronic phone. A short time later, there would be a warranty call on the whiz-bang phone, because it didn't handle humidity / grease / being dropped in a pot of chili nearly as well as the old phone. Once the PBX got analog rotary dial capability, the old phone was restored to the wall and everyone was happy.

Matt Roberds

Dr. Frankenstein has a cheap 'Abbynormal' brain waiting for you! ;-)

A carbon mic modulates the loop current. Always has, always will. That causes the 'C.O.' line voltage to drop whenever the diaphram is compressed by air pressure.

'Electronic' phones impose the AC voltae on the line without modulating the loop current directly.

If I remember correctly a phone line simulator is easy-peasy to make. If you're really, really attached to those old phones you could regenerate the power with a box that looks like a phone in one direction and a phone line (suitable for your phones) in the other.

But it'd be way less effort, and probably less cost, to just buy new phones.

--
My liberal friends think I'm a conservative kook. 
My conservative friends think I'm a liberal kook. 
Why am I not happy that they have found common ground? 

Tim Wescott, Communications, Control, Circuits & Software 
http://www.wescottdesign.com

What I was trying to say, and possibly not succeeding in saying, is that if the power supply at the "CO" (has a high impedance / has a high Thevenin equivalent resistance / isn't very "stiff") [pick your favorite term], there won't be as much loop current available, which will have an effect on the sound.

Here are some numbers to hopefully show what I am talking about. These are not meant to be too close to "real world" values - they are just to show the problem.

So, imagine as givens:

R1 Mic +---/\/\/\---o---/\/\/\---+

+| | V |

-| R2 | +---/\/\/\---o------------+

"CO" | Telephone set

The carbon microphone resistance Mic varies between 400 and 800 ohms as you talk. It does not vary based on any other factor.

There isn't any other resistance in the telephone set other than the carbon microphone.

The "good" CO is a 48 V DC voltage source with R1 = R2 = 400 ohm fixed resistor.

The "bad" CO is a 48 V DC voltage source with R1 = R2 = variable resistors that can be between 1600 and 3200 ohms.

The idea with the variable resistance is that it models a power supply that is not quite up to the job; something like a weak battery, or a filter capacitor that's way too small.

Calculate for the "good" CO:

With the carbon mike at 400 ohms, the loop current is

48 / (400 + 400 + 400) = 40 mA. The voltage at the telephone set will be (0.040 * 400) = 16 V.

With the carbon mike at 800 ohms, the loop current is

48 / (400 + 800 + 400) = 30 mA. The voltage at the telephone set will be (0.030 * 800) = 24 V.

So, for the "good" CO, the voltage at the telephone set changes from 16 to 24 V, with a loop current between 40 and 30 mA. The changes in voltage at the telephone set and loop current only depend on the change of resistance of the carbon microphone.

Calculate for the "bad" CO:

With the carbon mike at 400 ohms, the "bad" CO has a loop current of

48 / (1600 + 400 + 1600) = 13 mA to 48 / (3200 + 400 + 3200) = 7.1 mA. The voltage at the telephone set will be between (0.013 * 400) = 5.3 V and (0.0071 * 400) = 2.8 V.

With the carbon mike at 800 ohms, the "bad" CO has a loop current of

48 / (1600 + 800 + 1600) or 12 mA to 48 / (3200 + 800 + 3200) or 6.7 mA. The voltage at the telephone set will be between (0.012 * 800) or 9.6 V and (0.0067 * 800) or 5.3 V.

So, for the "bad" CO, the voltage at the telephone set changes from 2.8 to 9.6 V, with a loop current between 7.1 and 13 mA. The changes in voltage at the telephone set and loop current depend on the change of resistance of the carbon microphone *AND* the change in resistance at the CO.

The _higher_ resistance of the "bad" CO causes a lower amplitude signal to be produced by the microphone. If this was the only thing happening, the signal might sound OK, just not as loud as with the "good" CO.

The _changing_ resistance of the "bad" CO is adding more modulation to the loop current that you would expect from the microphone alone. This causes the resulting signal to be distorted and sound bad.

The higher and changing resistance is what I figured the Telo box is doing; that's why I said it might be "loaded down". As you said, the carbon microphone always loads down the CO. What I was trying to say is that if your CO is not very good, there may come a point where the normal load of the microphone, caused by talking, may be too much for the power supply in the CO; the power supply in the CO will start to sag even further than you would expect. That's all.

Matt Roberds

The loop current has to meet the specfication, or the call won't go hrough. As far as high impedance, they have 48 volt battery banks that deliver thousands of amps. All of this is prety much meaningless in ,ost ares, since they have small switching center scattered all over town, and no real C.O. Around here, it is fiber to the last mile, then a SLIC in a large outdoor cabinet to connect you over underground copper lines. How much drop do you see for two miles of AWG 24 or 22? Even though it's the last mile, you still have to meet the Bell standards if you want the line to work.

-snip-

But it seems clear that a modern electronic phone doesn't use a microphone that captures sound by varying resistance. Instead, the phone draws enough current for its own operation, and possibly some additional current to make sure the CO senses it as being off-hook. But then the audio signal is just injected onto what would otherwise be a steady DC voltage on the line. And what's needed to extract the audio signal at the CO or the Telo seems to be quite different for the two phone types.

But it's still curious to me how much the voltage varied with the old phone, while not at all with the new one, and that's on both Ma Bell and the Telo. The variation is far larger than would make any sense for an audio signal - volts instead of millivolts. Of course this is as measured with a digital meter - I wonder if an analog meter would show the same variation. By contrast, the digital meter shows a steady voltage even though the audio signal is actually there.

In any case, I'm going to be saving a bundle of money on the Ooma service over time, and I'm certainly not going to let the cost of three or four new phones (at maybe $15 each, and that's with Caller ID, redial, etc,) stand in the way - if that's what's needed, and it looks like it is.