Varistor Characteristics for Magnatone Vibrato Circuits

[Kagliostro]

The schematic repeat the values imprinted on the capacitors

47k is old notation and mean 0.047 (or 47000pf)

K

[martin manning]

47k is old notation and mean 0.047 (or 47000pf)

But kilo pico Farads is a bit confusing, and you'd actually have to write 47kpF to be precise... it's much clearer to use nano Farads, 47nF or just 47n.

[Kagliostro]

You are right ......

At the time I draw that schematic I used Jschem as CAD program for schematics, now I use FidocadJ but I'm thinking to move to Visio

I want to redraw that schematic in a better version, but I'm waiting to decide if use FidocadJ or start a new way with Visio

So I've think to correct the old notation on the new schematic, not to mess correcting the old one

K

[trobbins]

Arrgh, I miss-read your schematic Franco. The RC modulator arm comprises the 680pF cap from V3A. What is interesting is the DC coupling from the midpoint of the modulator arms (also the midpoint of the varistors) to the other mixer stage inputs of PI V6A stage.

V2B and V5A appear to have different bias conditions, and hence I would assume different anode idle voltage levels, which are possibly higher (?) than the varistor midpoint dc level. Those plate voltages at idle may then modify the DC current in the varistor string a bit, but I guess its not much given the 470K mixer resistances.

If you can measure the DC voltage at D' and bottom of varistor string when V3 is removed, then you should be able to deduce the voltage across each varistor and the current through the varistors, and hence compare with existing meassurements on the maggie varistors.

[Diablo1]

The schematic repeat the values imprinted on the capacitors

47k is old notation and mean 0.047 (or 47000pf)

K

OK, thanks.

[Kagliostro]

@ Trobbins

the amp is dismantled because there were problems with the PT wires and the OT was fried

when I'll assemble it I'll control the voltage

thanks

Franco

[Willabe]

As I understand it, the original parts were identified as:
Carborundum 233BNR-32
Workman FS1203 : 0.05mA @ 49V (Zenith 63-4906)
Workman FS605

New part is I believe:
Metrosil 100-P/W/921 : 1mA @ 100V; 0.1mA @ 55V

Others that have come up in discussions:
FR1039
FS-308
FS1211 : 0.05mA @ 61V (Zenith 63-5327)
FS1205 : 0.05mA @ 80V (Zenith 63-5058)

The original maggie circuit appears to be designed for a varistor conducting about 1mA at 100V, and about 45V at 0.05mA. Given that, I'm pretty sure the resulting effect with a slightly different varistor will be equivalent.

Hi all, I haven't posted before here and I hang out at Hoffman amps forum.

Great thread guys!

Someone there recently asked about where to buy varistors and as the thread went on, our mutual friend, Kagliostro posted a link to this thread. Thank's K!

My question is on the SiC varistor voltage rating. It seems Magnatone was using different voltage rated varistors in different amps?

The parts listed above are from 49-100v. I remember seeing somewhere recently you can use up to 150v varistor.

Also, I've seen some spec sheets list dcv and acv, might only be on MOV sheets? What voltage rating are we concerned with, acv or dcv? And in what voltage range can we use that will work in a Maggie circuit?

Did they adjust the LFO/driver signal/voltage to the varistors on different/lower voltage varistors?

The PI forces the voltage to cycle from idle condition of about 55-60V across each varistor, to the signal pk-pk limits of about 20-30V to about 90-100V (ie. up to a 70-80V pp swing). The varistor current never flows through zero, but sweeps between near 0.01mA to near 1mA.

I can see that trobbins knows what he's talking about by reading through this thread, as do many others here, but is R.G. Keen saying (somewhat) the same thing?

I attached a link below to R.G. Keen's site that has a few paragraphs on Maggie vibrato, scroll down to near the end.

R.G. writes Maggies are set up for 20-50v across the varistors.
http://geofex.com/Article_Folders/phasers/phase.html

After seeing NOS SiC varistors listed here at as low as 49v I bought a few off ebay, Russian SiC, rated at 56v, hoping they will work.

I also bought 4x NOS SiC from Warbler Muse to A/B against the replacement you all have been working on and the Russians.

I'm working on a bread board for testing.


Thanks, Brad

[trobbins]

As I understand it, Magnatone used effectively the same varistor type. And has been noted, those parts can vary in tested performance by at least +/-10%. The maggie circuits dictate the operating conditions for the varistors, given that mains voltage and part variations easily contribute at least +/-20% tolerance on any particular voltage-current operating point.

The spec sheets for those old parts are on-line, as are some test results, and show voltage versus current. AC voltage is just a varying dc voltage.

There are a few variations of circuits that Magnatone used - I haven't analysed them all, but I expect they generally operate the varistors in a similar manner.

RGK didn't say what you have stated - perhaps you should re-read his discussion again.

Anyone can go about redesigning the maggie circuits if they feel inclined to see if that achieves similar performance with different varistor types. The basic design information was given by Moses, and the rest is relatively simple circuit design and testing, and of course time and incentive.

Ciao, Tim

[Willabe]

The spec sheets for those old parts are on-line, as are some test results, and show voltage versus current.

Thanks, I'll do a search for them.

AC voltage is just a varying dc voltage.

Yes, but the voltages listed here for NOS varistors doesn't say acv or dcv.

Are you saying it doesn't matter here for the purpose their being used?

RGK didn't say what you have stated - perhaps you should re-read his discussion again.

I'm sorry I was trying to ask that as a question. ("but is R.G. Keen saying (somewhat) the same thing?")

Sorry, but again I don't understand. (I did read it again.)

You post; "20-30V to about 90-100V (ie. up to a 70-80V pp swing)."

R.G. Keen;
"Varistors had a resistance that was variable, depending on how much voltage was across them. The higher the voltage, the lower the resistance they showed to a signal. The Varistors of the '50's were not very sensitive, and it took 20-50V across them to make a difference in the resistance."

I must be not understanding this correctly. I see 70-80V pp swing across the varistor and only 20-50V(pp swing?) across the varistors from R.G.?

I'm just trying to understand how much voltage swing (which would have to be ACV) do they need across them to work?

How can a 70-80V pp or 20-50V pp across a 49v varistor not burn them up,
ie, Workman FS1203 : 0.05mA @ 49V (Zenith 63-4906)?

I'm sure I'm not seeing/thinking about this correctly. Any help untangling this in my mind will be greatly appreciated.

I'll look for what Moses posted, thank you.

Brad

[trobbins]

Moses article assists in defining the phase shift achieved from a given change in RC time constant (which changes due to sweep in R obtained from the varistors as the LFO signal changes their operating voltage across a range of eg. 30-100V).

You could use a spice sim to generate the same results, but poor old Moses had to number crunch his results no doubt.

So you have to assess each varistor for its change in resistance versus change in operating voltage, and design a circuit that sweeps the varistor voltage sufficiently to get the aimed for change in varistor resistance, that will then give the hoped change in phase shift, which your ears will then be pleased with.

The maggie circuit has gone down that design path for you. In some amps they chose to 'double' the phase shift swing by using two sequential stages. I reckon one stage sounds great.

PS. varistors are often defined by a certain voltage-current point. That is not a max or min or sweet spot or ... rating, its just a snapshot which may or may not aid in comparing to other varistors. Some of the datasheets also plot a power dissipation limit, which would be a limit for the device not to go beyond. The datasheets show the characteristic curve of V versus I, which is needed for deign of the surrounding circuitry.

[Willabe]

Thanks, that helps.

Brad

[Willabe]

I did some searching for the article that Moses wrote but all I found was a 2007 post at another forum from Merlin where he mentions it.

And I think his 1st name was Richard and he wrote it in 1953?

Does anyone know the title of the article?

Thanks, Brad

[Willabe]

Stephen Keller has the reference on his informative website http://thermionic.info/

Kagliostro showed me where to find it. Thank You K.

"Phase Angle Measurements in A.F."

And;

Richard Dorf, "The Universal Vibrato"

Richard Dorf, "The Wurlitzer Vibrato"

Thank you, Brad

[trobbins]

I'm preparing a draft doc that summaries the vibrato circuit technique and design, and provides links to relevant docs - that should help the learning curve. I may get it uploaded over the weekend.

[Willabe]

That will be great trobbins!

Could you include some information on how you guys test varistors to match them and what is considered a matched pair?

From what you said a couple of posts back, I was thinking that because the resistance of the varistor changes with the voltage across them, they would have to be measured for resistance at 2 different voltages, top of the LFO swing and bottom of the LFO swing?

So that a pair, 1 driven from the LFO/drivers plate, the other driven from the LFO/drivers K would track together equally pp to keep the AC sine wave symmetrical?

Brad