VOIP, ATA REN and old analog phones ring voltage booster

[CapLeaker]

I am working on a backup phone to VHF paging solution. Basically I am dealing with the old Zetron pagers. Here is how it works: One calls a number, the Zetron picks up, operator punches in the DTMF access code, the Zetron keys the transmitter and you can talk. press # the Zetron unkeys the transmitter and hangs up / times out.
Now this setup works good on the old land lines, but not at all using a VOIP ATA. You can call all you want and swear like a bastard at the Zetron, but it will not (even if I threaten it’s life) pick up.
Yep, I know about the high ring voltage setting in the setup menu in certain ATA’s, but that’s still has not enough OOOMPH.
I thought about modding the Zetron to detect lower ring voltage, but that means it’s getting complicated, because I’d have to have a switch, so I can choose to operate the Zetron between a regular land line and an ATA. Plus I gotta do that on all the Zetrons. The ring voltage here is 90V AC, 20Hz.
I found something like a ring voltage booster made by Viking Electronics, Model RG-10a. A bit pricey.
I also found another one called a “Ring Voltage Booster II”.
Did any one deal with this 90V AC ring voltage problem before or has a better, easier and cheaper, maybe even a home brew solution, diagram?

[redwire]

Are you sure it's ring voltage the Zetron has issues with?
I'm not sure all ATA's are isolated. The Zetron might not be isolated from the phone line as well, so a ground fault or floating tip+ring might mess it up. You'd have to force it pick up and see if the audio is working OK.

Ring detect circuits are very similar - an opto-coupler and blocking cap and zener(s) or resistor divider. You could lower the Zetron threshold, assuming it is too low. Or maybe the cap/opto is weak due to age or lightning.

[CapLeaker]

I extracted the schematic out of the manual... I should have done this the first place, but it was getting late last night.
Yes, these Zetron's are a bit dated, but replacing all of them for newer model isn't going to happen. They are quite bullet proof and yes, I did check the input protection to see what's going on there, but it was fine. I am not much of a phone guy, but I do know that much that the ring voltage out of most ATA's is not 90V AC. Also with ATA I mean a simple box like a Granstream etc, not some card to stick in a fancy phone system.
Modding the ring detection circuit by changing the zener diode CR17 may be worth a try, it may also need lower resistor values in R30 and R31.
Problem with this idea is that I have to go to each unit, put a temporary Zetron in that needs to be programmed same way. Not so much of a problem using the TL-866II and using an EEPROM rather than an EPROM plus a spare battery backed up S-RAM (because then I wouldn't need a stack of EPROMs or bring an UV eraser with me too). Clone the EPROM, clone the S-RAM, install the temporary Zetron. Then mod the original unit as such that the new modification is switchable or auto switch, so both systems can be used. Then go to each spot and change these Zetron's out again in order to put the original one back. There is only one spare Zetron, so I can't go and switch out a dozen at once. That's would keep me busy for a long time and is a pain in the doin it that way... just for a backup solution. But I have been surprised before.

[stj]

a DTMF gateway - nice.
i was going to build one until radio shack died.
then i had no easy source of cheap dtmf decoder chips.
i wasnt going to key a transmitter, i was going to control a relay-bank.

[CapLeaker]

If still got DTMF decoder chips from repairing TNC’s. There was a year, where I repaired piles of them. Also got the old Zylog CPUs and SIO, RAM and other chips. A home brew relay controller you say using DTMF decoders? Sounds interesting!

[stj]

for turning stuff on/off while out - smart home before cellfones

[redwire]

Your problem still doesn't make sense. If it is a wimpy ATA, most are rated REN 1, which is one mechanical bell say 65mA (FCC 68 1,400Ω 90VAC). Way more current than an opto-coupler LED or modern phone ringer. You'd have to measure what the ATA is putting out under load. When I was working in telephony, the ATA ringer power is a major expense so they want that as small and wimpy as possible.

Long time ago I got a contract making a DTMF switched relay, one was for temporarily shutting off grocery warehouse coolers during peak power rates, and the other was for ATM's to activate some backup comms I think. Used a PIC 12C509 and Mitel MT8870.

edit: small chance but Caller ID packet might be confusing the Zetron as well. SLIC LE9643 (see 8.5) uses a tiny SOT-23 boost converter to make 50VAC into REN 5, and 60VAC into REN 3, from a 9-15VDC source. It's not hard to make enough ring power.

[CapLeaker]

Redwire, I think that you may just have explained it whats going on. Maybe the Zetron is putting on a to high of load to the ATA. Like I said, I am not really a phone guy. I used to replace the input protection on various phone gear, repair other phone systems etc, but never had something to do with ring voltage, loop current etc.
Around here ATM's don't have backup comms. How ever if an ATM looses comms or any other dinky problem, someone gets sent there to investigate. I used to work for such an outfit 20 years ago.

Well, if you go at the REN rating from an ATA, I am not so sure if that is the same REN ratings on the old stuff. All I remember is seeing anywhere between 75VAC to 100VAC around here as ring voltage on the land lines. These days most everything is getting swapped over to VOIP. Getting hard pressed finding a land line now. Even at my work it's all VOIP. There is only one place where I know there is still a landline that I have access to.

[redwire]

If you can, measure the ring voltage going into the Zetron. Some multimeters will not read properly on ACV when there is DCV present. At least say 50VAC should be enough to be a ring, unless the Zetron has some hardware problems. There are transients and lightning that can zap the ring detect capacitor, but I find then a device will just go and stay off-hook when connected, a steady DC load does that.
Some better analog phone tech meets the FCC 600VAC surge spec using MOV's or gas-tubes, some PTC's.

You could post pics of it, to see how the circuit does ring detect and why that is finnicky to an ATA. It could be a grounding issue too, if the Zetron is earth-grounded and the ATA too then tip/ring polarity matters.

[CapLeaker]

It has a couple of MOVs first thing on the input and those are fine. I will have to bring everything to my bench at home, then I can fiddle with it around better. As for multimeters, I use usually a Fuke 279 or a 289 and the later one gets a yearly calibration (work requirement). I did post a schematic of it, but I will upload some pics when I get it home.

[redwire]

I missed the schematic - it's your typical ring detect circuit, about 0.1 REN load and needing at least say ~30VAC min. to light the opto-coupler LED. An ATA should drive it no prob. RING also activates the front panel LED. It looks like there is no battery to store a config such as # rings.
So I'm wondering if it is the relay or something else not working.
Fluke multimeters are AC-coupled so they don't get messed up with DC present.

[CapLeaker]

I'll be getting all this stuff home and set it up by tomorrow. The ring detect is nothing fancy, but my ATA will not handle it. The ATA works fine with the newish analog dial pad style phone. I also arranged for an old phone with a real bell. Maybe the Zetron is good and my ATA is bad?

[redwire]

Maybe there is a grounding problem? ATA's don't have floating or isolated outputs, ring is usually earth ground. Not sure what power adapter it has.
The Dex has isolated line interface though unless some jumper is in? Not sure what power adapter it has.
Just to keep in mind if things get weird.

[CapLeaker]

Things are weird all right. So here is what I got.
ATA = Grandstream HT701
1 old rotary pulse dial phone with a bell ringer
1 analog wireless DECT phone
1 Zetron Model 32 DAPT-JR rev C board.

Right now I have a line splitter direct in the ATA and hooked up the old rotary phone and the wireless DECT phone. Guess what? Works like a charm! Bells ring, no problem!

As soon as I plug into that Zetron by itself or paired up, things are not so good. The Ring indicator only gives short blips, but it will not for its own life pick up. For example, I paired up the old rotary phone with the Zetron. No go, the ole phone wont even ring. As soon as I unplug the Zetron, it works perfect.

It's got to be something with the Zetron.

[CapLeaker]

For F*CK sake's. The problem is or was in the Zetron all right. I went all over that damned heap of pieces of sh*t. Blah... checked the MOVs, the diodes, switching ring and tip, changing PSU (wrong PSU with it), nothing worked.

Look over that big yellow transformer. There is a Matrix board. All I did was to move that board over ONE NOTCH! So the setting is E-E. That's it!

Stupid thing works perfect now. No ring booster needed anymore!

[stj]

fuse/holder corroded badly,
the 10uf cap next to it is expelling it's bung!

[CapLeaker]

Whats the difference between DID, E-E and E+M?
Direct In Dial
End to End
End + ?
Any old phone guys around here?

In the DID setting of the matrix, there would be nothing connected between either sides (open).
In the E-E setting there is a 150 Ohm 1 watt resistor across A2 and R1 on the matrix
In the E+M setting there is a jumper between A2 and A1 on the matrix.

Thanks stj... I saw that too... but that is a way too easy fix!

[redwire]

Zetron data
Zetrom Model 32 DAPT-Jr Manual 189 pages of fun.
Lots of programming and I hope the M48Z08 battery ain't dead, date code '99.
From the manual, the Matrix plug:
DID = Telco is DID/PBX Loop/Local
E&M = Telco is E&M Type I tie-trunk
E-E = Telco is End-to-End Phone Line

[CapLeaker]

I did copy already the EPROM and the RAM with the battery inside.
Went and got a phone ring generator. It’s 100VAC @ 30Hz. To make a sanity check.
Phone ring generator, splitter, Zetron and the old rotary phone.
Set the Zetron matrix to DID: Zetron sees the ringer, phone doesn’t ring and the ring generator sounds like it has a load on it.
Set the matrix to E-E: everything is working and happy as it should.
Now if someone can explain that in simple English why that is, I’d greatly appreciate it. Guess I got a 200 paper toilet read to do.
Question is which setting is correct? I mean the Zetron's are just on a stinking regular phone line. It's not going through some crazy company phone network. So IMHO the end to end setting is correct?

[CapLeaker]

Verdict: the e-e (end to end) is the correct one. The matrix changes the tap points on the transformer, to accommodate what ever other phone line system works. The DID setting will work on a true end to end phone line, but it’s not correct as other phones connected to the same line won’t ring. However the Zetron will see the ring and pick up. The problem never surfaced for like 25 to 30 years, because the Zetron was on a single dedicated phone line.