diyAudio (
-   Pass Labs (
-   -   F5 measurement/tuning (

doolav 11th December 2012 12:27 PM

F5 measurement/tuning
Hi all,

I recently built a clone of F5 (stock, rev. 1, dual mono, 250VA transformers/8x15mF/23.7V per channel).
It's still has a few hours of burning in time, but sounds (sounded :D) well.
Not yet much experience in electronics ("can read with a dictionary"), hence some dummy questions:
currently I am trying to measure its square wave response, and it does not look very impressive (miles away from the nice green 200kHz pic in the manual on the firstwatt site): one channel has very much rounded corners, the other has a significant overshoot on top half and some undershoot on the bottom half (sorry, no pictures yet - will try to take them whenever I can). This is the same for different frequencies - 10k, 50k, 100k. The input signal amplitude is 1Vpp from a Tek AFG3102 signal gen, and it looks good when fed directly into the scope (Tek TDS320).
I tried it without a load and with 6ohm speakers connected (don't have a proper resistor for load), I have tried to rotate transformer, move input wires a bit - no change.
What can be wrong here - bad layout/wiring/resistors matching?
What amplitude of the test signal is safe to feed into F5? My understanding was it's about +-2.5V max, is that correct?

Yesterday I tried to test it with an innocent 2.5kHz square wave (again, 1Vpp), this gave me bad oscillation on the overshooted/undershooted channel, I even noticed HF noise coming from the amp (no speakers connected), then something started to smoke (very slightly), and I noticed the signal amplitude on the scope started to compress a bit, and I immediately shut it off. Switched it on later without any input, no smoke, but did not check yet if it's alive or not. Could not identify visually if anything has been burnt - no colour change or anything else.

Really appreciate any advise from more experienced of you than me :confused:

PS I did search quite a lot prior to post here, but still did not find or recognize the answers

AndrewT 11th December 2012 12:36 PM

check output offset.
check output bias.

connect to your source (wave generator) with generator output turned down to ZERO.
check output offset.
check output bias.

Turn up the 1kHz (or close) signal to 10mVac.
Measure the F5 output waveform. use both a DMM set to 2.000Vac and use your scope to measure the peak to peak. Expect about 180mVpp for 60mVac.

Add a standard 22r across the output. The power will be ~ 0.14mW
Add a second 22r for an 11r loading.
Add a third 22r for a 7r3 loading.
Slowing turn up the gen signal from 10mVac to 100mVac. Power in each resistor will rise to 16mW. They will still be cold.
switch to different frequencies and check the wave shape stays same or similar.

doolav 11th December 2012 12:44 PM

Hi Andrew,

the 22r - what power it should be?

Also, why having output loaded while measuring is necessary? (or even apart from measuring - does it mean my amp is unstable or it's not meant to work without a load?)


AndrewT 11th December 2012 01:32 PM

a standard 1/4W or 1/2W resistor.
I happen to have 600mW because we have them here in Europe and they are cheap.

Some amplifiers do not like operating without some load.
Amplifiers that have no Thiele Network, nor output Zobel have no load. These may particularly require a little bit of loading.

Oops, that word should be slowly !

doolav 11th December 2012 01:37 PM

Thanks Andrew,

here is quite a bit to try for me.

So far:
offset (while setting the bias, not while testing with sig gen) is ~ +-5mV
bias ~1.3A (~880mV across r68)

"Some amplifiers do not like operating without some load"
- having a 6 Ohm speaker is not considered as a load?

will get and post those with sq. wave in later


Andersonix 13th December 2012 02:44 AM

How close are the gate blocker resistors to the gates? Can we see pics of the general layout? Andrew likes large pics...

AndrewT 13th December 2012 12:07 PM

Andrew T likes large enough pictures to see the detail that the poster requires us to see.

That automatically rules out 10Mpixel to 20Mpixel pics that are out of focus and also rules out the uncropped surroundings of carpets, wallpaper, views out the window, etc.

h_a 14th December 2012 07:09 AM

Well, commonly signal generators assume a 50 Ohms load and for that load they're calibrated.

So, unless your signal generator is a different type, you need a 50 Ohm resistor (higher wattage type for serious signal) in parallel to the amp input; also I recommend koaxial 50 Ohms wiring from the generator to the input.


EUVL 14th December 2012 11:16 AM

> Well, commonly signal generators assume a 50 Ohms load and for that load they're calibrated.

Really ?
They must then all have a beefy output stage with TO220 devices .....


AndrewT 14th December 2012 11:25 AM

Indeed they have good current capability.
Both my generators are 50ohm. That requires an internal 50r in the feed to the output BNC, then 50ohm coax to the load and finally 50r load across the coax output.

I think mine have a 5Vac max output (into 50r). But that is still only 1/4W into the 100r load that the output stage sees.

All times are GMT. The time now is 08:03 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2