]> sigrok.org Git - libsigrok.git/commitdiff
hantek-dso: calculate voltage based on vdiv setting
authorBert Vermeulen <redacted>
Tue, 5 Jun 2012 16:17:01 +0000 (18:17 +0200)
committerBert Vermeulen <redacted>
Tue, 5 Jun 2012 16:18:23 +0000 (18:18 +0200)
hardware/hantek-dso/api.c

index 5f2301ff9167eff0ac7fe48e5f038dc0dc05df9e..c571caf89ee92d258886ea547b8d703654fa023f 100644 (file)
@@ -555,7 +555,7 @@ static void receive_transfer(struct libusb_transfer *transfer)
        struct sr_datafeed_packet packet;
        struct sr_datafeed_analog analog;
        struct context *ctx;
-       float ch1, ch2;
+       float ch1, ch2, range;
        int num_probes, data_offset, i;
 
        ctx = transfer->user_data;
@@ -581,14 +581,24 @@ static void receive_transfer(struct libusb_transfer *transfer)
                /* The device always sends data for both channels. If a channel
                 * is disabled, it contains a copy of the enabled channel's
                 * data. However, we only send the requested channels to the bus.
+                *
+                * Voltage values are encoded as a value 0-255 (0-512 on the 5200*),
+                * where the value is a point in the range represented by the vdiv
+                * setting. There are 8 vertical divs, so e.g. 500mV/div represents
+                * 4V peak-to-peak where 0 = -2V and 255 = +2V.
                 */
                /* TODO: support for 5xxx series 9-bit samples */
                if (ctx->ch1_enabled) {
-                       ch1 = (*(transfer->buffer + i * 2 + 1) / 255.0);
+                       range = ((float)vdivs[ctx->voltage_ch1].p / vdivs[ctx->voltage_ch1].q) * 8;
+                       ch1 = range / 255 * *(transfer->buffer + i * 2 + 1);
+                       /* Value is centered around 0V. */
+                       ch1 -= range / 2;
                        analog.data[data_offset++] = ch1;
                }
                if (ctx->ch2_enabled) {
-                       ch2 = (*(transfer->buffer + i * 2) / 255.0);
+                       range = ((float)vdivs[ctx->voltage_ch2].p / vdivs[ctx->voltage_ch2].q) * 8;
+                       ch2 = range / 255 * *(transfer->buffer + i * 2);
+                       ch2 -= range / 2;
                        analog.data[data_offset++] = ch2;
                }
        }